vultr/openclaw
1
0
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-05-06 09:00:42 +00:00
Code Issues Packages Projects Releases Wiki Activity

test: show bootstrap context in prompt snapshots (#76302)

Browse Source
This commit is contained in:
pashpashpash
2026-05-02 16:29:34 -07:00
committed by GitHub
parent a1054fbe1b
commit ad0e1a0d5d
6 changed files with 215 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
test/fixtures/agents/prompt-snapshots/codex-runtime-happy-path/README.md vendored
View File

@@ -8,7 +8,9 @@ These fixtures capture the default OpenAI/Codex happy path for prompt review:
- `messages.visibleReplies: "message_tool"`, which is the Codex-harness default for visible source replies.
- Telegram direct chat, Discord group chat, and a heartbeat turn with `heartbeat_respond` available.
The Markdown files show selected app-server thread/turn params plus a reconstructed model-bound prompt layer stack: Codex `gpt-5.5` model instructions from a pinned Codex model catalog fixture, Codex permission developer instructions for the happy-path yolo profile, OpenClaw developer instructions, user turn input, and references to the complete dynamic tool catalog.
The Markdown files show selected app-server thread/turn params plus a reconstructed model-bound prompt layer stack: Codex `gpt-5.5` model instructions from a pinned Codex model catalog fixture, Codex permission developer instructions for the happy-path yolo profile, simulated OpenClaw workspace bootstrap config instructions, OpenClaw developer instructions, user turn input, and references to the complete dynamic tool catalog.
The workspace bootstrap simulation includes dummy `SOUL.md`, `TOOLS.md`, and `HEARTBEAT.md` contents so prompt reviewers can see how those OpenClaw project/user context files are forwarded to Codex. `AGENTS.md` is intentionally not repeated here because Codex loads it natively.
The tool catalog is pinned to the canonical happy-path OpenClaw tools so optional locally installed plugin tools do not create fixture churn.
@@ -18,7 +20,7 @@ The Codex model prompt fixture is generated from the same Codex model catalog/ca
pnpm prompt:snapshots:sync-codex-model
```
These snapshots are still not a byte-for-byte raw OpenAI request capture. Codex-owned workspace context such as `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions can be added inside the Codex runtime after OpenClaw sends thread and turn params.
These snapshots are still not a byte-for-byte raw OpenAI request capture. Codex-owned native `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions can be added inside the Codex runtime after OpenClaw sends thread and turn params.
Regenerate with:

52
test/fixtures/agents/prompt-snapshots/codex-runtime-happy-path/discord-group-codex-message-tool.md vendored
View File

@@ -7,6 +7,7 @@
- Default happy path: the same Codex agent is mentioned in a Discord group/channel while Telegram can remain the user's primary direct interface.
- Group-visible output must be explicit through the message tool; the model is also told to mostly lurk unless directly addressed or clearly useful.
- This captures the OpenClaw-owned Codex app-server inputs and reconstructs the stable Codex model/permission layers from committed Codex prompt fixtures.
- This also simulates workspace bootstrap files forwarded through Codex `config.instructions`: `SOUL.md`, `TOOLS.md`, and `HEARTBEAT.md`.
## Scenario Metadata
@@ -19,6 +20,11 @@
"model": "gpt-5.5",
"modelProvider": "openai",
"runtime": "codex_app_server",
"simulatedWorkspaceBootstrapFiles": [
"/tmp/openclaw-happy-path/workspace/SOUL.md",
"/tmp/openclaw-happy-path/workspace/TOOLS.md",
"/tmp/openclaw-happy-path/workspace/HEARTBEAT.md"
],
"sourceReplyDeliveryMode": "message_tool_only",
"toolSnapshot": "codex-dynamic-tools.discord-group.json",
"trigger": "user"
@@ -69,6 +75,9 @@
{
"approvalPolicy": "never",
"approvalsReviewer": "user",
"config": {
"instructions": "OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.\n\n# Project Context\n\nThe following project context files have been loaded:\nIf SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.\n\n## /tmp/openclaw-happy-path/workspace/SOUL.md\n\n<SOUL.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/TOOLS.md\n\n<TOOLS.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md\n\n<HEARTBEAT.md contents will be here>"
},
"cwd": "/tmp/openclaw-happy-path/workspace",
"developerInstructions": "<see Reconstructed Model-Bound Prompt Layers>",
"dynamicTools": [
@@ -103,6 +112,9 @@
{
"approvalPolicy": "never",
"approvalsReviewer": "user",
"config": {
"instructions": "OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.\n\n# Project Context\n\nThe following project context files have been loaded:\nIf SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.\n\n## /tmp/openclaw-happy-path/workspace/SOUL.md\n\n<SOUL.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/TOOLS.md\n\n<TOOLS.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md\n\n<HEARTBEAT.md contents will be here>"
},
"developerInstructions": "<see Reconstructed Model-Bound Prompt Layers>",
"model": "gpt-5.5",
"persistExtendedHistory": true,
@@ -136,7 +148,7 @@
## Reconstructed Model-Bound Prompt Layers
This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.
This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, simulated OpenClaw workspace bootstrap config instructions, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as native workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.
### Layer Metadata
@@ -161,9 +173,10 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
},
"limitations": [
"This is a reconstructed prompt-layer snapshot, not a byte-for-byte raw OpenAI request captured from Codex core.",
"Codex-owned workspace and app context is listed as a runtime-owned gap until Codex exposes a rendered-prompt inspection API."
"Codex-owned workspace AGENTS.md, environment context, memories, app/plugin instructions, and provider tool serialization are still runtime-owned gaps until Codex exposes a rendered-prompt inspection API."
],
"openClawRuntime": {
"configInstructionsFrom": "extensions/codex app-server thread/start config.instructions",
"developerInstructionsFrom": "extensions/codex app-server thread/start developerInstructions",
"dynamicToolsFrom": "codex-dynamic-tools.discord-group.json",
"userInputFrom": "extensions/codex app-server turn/start input"
@@ -183,6 +196,10 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
"chars": 307,
"roughTokens": 77
},
"codexWorkspaceBootstrapConfigInstructions": {
"chars": 632,
"roughTokens": 158
},
"dynamicToolsJson": {
"chars": 50058,
"roughTokens": 12515
@@ -192,12 +209,12 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
"roughTokens": 2151
},
"totalTextOnly": {
"chars": 31122,
"roughTokens": 7781
"chars": 31756,
"roughTokens": 7939
},
"totalWithDynamicToolsJson": {
"chars": 81182,
"roughTokens": 20296
"chars": 81816,
"roughTokens": 20454
},
"userInputText": {
"chars": 870,
@@ -373,6 +390,29 @@ Filesystem sandboxing defines which files can be read or written. `sandbox_mode`
Approval policy is currently never. Do not provide the `sandbox_permissions` for any reason, commands will be rejected.
```
### User: Codex Config Instructions (OpenClaw Workspace Bootstrap Context)
```text
OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.
# Project Context
The following project context files have been loaded:
If SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.
## /tmp/openclaw-happy-path/workspace/SOUL.md
<SOUL.md contents will be here>
## /tmp/openclaw-happy-path/workspace/TOOLS.md
<TOOLS.md contents will be here>
## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md
<HEARTBEAT.md contents will be here>
```
### Developer: OpenClaw Runtime Instructions
````text

52
test/fixtures/agents/prompt-snapshots/codex-runtime-happy-path/telegram-direct-codex-message-tool.md vendored
View File

@@ -7,6 +7,7 @@
- Default happy path: OpenAI model through the Codex harness/runtime, Telegram direct conversation, and message-tool-only visible replies.
- A quiet turn is represented by not calling `message(action=send)`; the normal final assistant text is private to OpenClaw/Codex.
- This captures the OpenClaw-owned Codex app-server inputs and reconstructs the stable Codex model/permission layers from committed Codex prompt fixtures.
- This also simulates workspace bootstrap files forwarded through Codex `config.instructions`: `SOUL.md`, `TOOLS.md`, and `HEARTBEAT.md`.
## Scenario Metadata
@@ -19,6 +20,11 @@
"model": "gpt-5.5",
"modelProvider": "openai",
"runtime": "codex_app_server",
"simulatedWorkspaceBootstrapFiles": [
"/tmp/openclaw-happy-path/workspace/SOUL.md",
"/tmp/openclaw-happy-path/workspace/TOOLS.md",
"/tmp/openclaw-happy-path/workspace/HEARTBEAT.md"
],
"sourceReplyDeliveryMode": "message_tool_only",
"toolSnapshot": "codex-dynamic-tools.telegram-direct.json",
"trigger": "user"
@@ -69,6 +75,9 @@
{
"approvalPolicy": "never",
"approvalsReviewer": "user",
"config": {
"instructions": "OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.\n\n# Project Context\n\nThe following project context files have been loaded:\nIf SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.\n\n## /tmp/openclaw-happy-path/workspace/SOUL.md\n\n<SOUL.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/TOOLS.md\n\n<TOOLS.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md\n\n<HEARTBEAT.md contents will be here>"
},
"cwd": "/tmp/openclaw-happy-path/workspace",
"developerInstructions": "<see Reconstructed Model-Bound Prompt Layers>",
"dynamicTools": [
@@ -103,6 +112,9 @@
{
"approvalPolicy": "never",
"approvalsReviewer": "user",
"config": {
"instructions": "OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.\n\n# Project Context\n\nThe following project context files have been loaded:\nIf SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.\n\n## /tmp/openclaw-happy-path/workspace/SOUL.md\n\n<SOUL.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/TOOLS.md\n\n<TOOLS.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md\n\n<HEARTBEAT.md contents will be here>"
},
"developerInstructions": "<see Reconstructed Model-Bound Prompt Layers>",
"model": "gpt-5.5",
"persistExtendedHistory": true,
@@ -136,7 +148,7 @@
## Reconstructed Model-Bound Prompt Layers
This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.
This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, simulated OpenClaw workspace bootstrap config instructions, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as native workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.
### Layer Metadata
@@ -161,9 +173,10 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
},
"limitations": [
"This is a reconstructed prompt-layer snapshot, not a byte-for-byte raw OpenAI request captured from Codex core.",
"Codex-owned workspace and app context is listed as a runtime-owned gap until Codex exposes a rendered-prompt inspection API."
"Codex-owned workspace AGENTS.md, environment context, memories, app/plugin instructions, and provider tool serialization are still runtime-owned gaps until Codex exposes a rendered-prompt inspection API."
],
"openClawRuntime": {
"configInstructionsFrom": "extensions/codex app-server thread/start config.instructions",
"developerInstructionsFrom": "extensions/codex app-server thread/start developerInstructions",
"dynamicToolsFrom": "codex-dynamic-tools.telegram-direct.json",
"userInputFrom": "extensions/codex app-server turn/start input"
@@ -183,6 +196,10 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
"chars": 307,
"roughTokens": 77
},
"codexWorkspaceBootstrapConfigInstructions": {
"chars": 632,
"roughTokens": 158
},
"dynamicToolsJson": {
"chars": 49749,
"roughTokens": 12438
@@ -192,12 +209,12 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
"roughTokens": 1934
},
"totalTextOnly": {
"chars": 29751,
"roughTokens": 7438
"chars": 30385,
"roughTokens": 7597
},
"totalWithDynamicToolsJson": {
"chars": 79502,
"roughTokens": 19876
"chars": 80136,
"roughTokens": 20034
},
"userInputText": {
"chars": 370,
@@ -373,6 +390,29 @@ Filesystem sandboxing defines which files can be read or written. `sandbox_mode`
Approval policy is currently never. Do not provide the `sandbox_permissions` for any reason, commands will be rejected.
```
### User: Codex Config Instructions (OpenClaw Workspace Bootstrap Context)
```text
OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.
# Project Context
The following project context files have been loaded:
If SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.
## /tmp/openclaw-happy-path/workspace/SOUL.md
<SOUL.md contents will be here>
## /tmp/openclaw-happy-path/workspace/TOOLS.md
<TOOLS.md contents will be here>
## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md
<HEARTBEAT.md contents will be here>
```
### Developer: OpenClaw Runtime Instructions
````text

52
test/fixtures/agents/prompt-snapshots/codex-runtime-happy-path/telegram-heartbeat-codex-tool.md vendored
View File

@@ -7,6 +7,7 @@
- Heartbeat happy path: Codex receives the structured `heartbeat_respond` dynamic tool because `messages.visibleReplies` is `message_tool`.
- The heartbeat tool carries the notify/no-notify decision, outcome, summary, and optional notification text instead of relying only on final-text parsing.
- This captures the OpenClaw-owned Codex app-server inputs and reconstructs the stable Codex model/permission layers from committed Codex prompt fixtures.
- This also simulates workspace bootstrap files forwarded through Codex `config.instructions`: `SOUL.md`, `TOOLS.md`, and `HEARTBEAT.md`.
## Scenario Metadata
@@ -19,6 +20,11 @@
"model": "gpt-5.5",
"modelProvider": "openai",
"runtime": "codex_app_server",
"simulatedWorkspaceBootstrapFiles": [
"/tmp/openclaw-happy-path/workspace/SOUL.md",
"/tmp/openclaw-happy-path/workspace/TOOLS.md",
"/tmp/openclaw-happy-path/workspace/HEARTBEAT.md"
],
"sourceReplyDeliveryMode": "message_tool_only",
"toolSnapshot": "codex-dynamic-tools.heartbeat-turn.json",
"trigger": "heartbeat"
@@ -69,6 +75,9 @@
{
"approvalPolicy": "never",
"approvalsReviewer": "user",
"config": {
"instructions": "OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.\n\n# Project Context\n\nThe following project context files have been loaded:\nIf SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.\n\n## /tmp/openclaw-happy-path/workspace/SOUL.md\n\n<SOUL.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/TOOLS.md\n\n<TOOLS.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md\n\n<HEARTBEAT.md contents will be here>"
},
"cwd": "/tmp/openclaw-happy-path/workspace",
"developerInstructions": "<see Reconstructed Model-Bound Prompt Layers>",
"dynamicTools": [
@@ -104,6 +113,9 @@
{
"approvalPolicy": "never",
"approvalsReviewer": "user",
"config": {
"instructions": "OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.\n\n# Project Context\n\nThe following project context files have been loaded:\nIf SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.\n\n## /tmp/openclaw-happy-path/workspace/SOUL.md\n\n<SOUL.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/TOOLS.md\n\n<TOOLS.md contents will be here>\n\n## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md\n\n<HEARTBEAT.md contents will be here>"
},
"developerInstructions": "<see Reconstructed Model-Bound Prompt Layers>",
"model": "gpt-5.5",
"persistExtendedHistory": true,
@@ -137,7 +149,7 @@
## Reconstructed Model-Bound Prompt Layers
This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.
This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, simulated OpenClaw workspace bootstrap config instructions, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as native workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.
### Layer Metadata
@@ -162,9 +174,10 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
},
"limitations": [
"This is a reconstructed prompt-layer snapshot, not a byte-for-byte raw OpenAI request captured from Codex core.",
"Codex-owned workspace and app context is listed as a runtime-owned gap until Codex exposes a rendered-prompt inspection API."
"Codex-owned workspace AGENTS.md, environment context, memories, app/plugin instructions, and provider tool serialization are still runtime-owned gaps until Codex exposes a rendered-prompt inspection API."
],
"openClawRuntime": {
"configInstructionsFrom": "extensions/codex app-server thread/start config.instructions",
"developerInstructionsFrom": "extensions/codex app-server thread/start developerInstructions",
"dynamicToolsFrom": "codex-dynamic-tools.heartbeat-turn.json",
"userInputFrom": "extensions/codex app-server turn/start input"
@@ -184,6 +197,10 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
"chars": 307,
"roughTokens": 77
},
"codexWorkspaceBootstrapConfigInstructions": {
"chars": 632,
"roughTokens": 158
},
"dynamicToolsJson": {
"chars": 50872,
"roughTokens": 12718
@@ -193,12 +210,12 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
"roughTokens": 1934
},
"totalTextOnly": {
"chars": 29849,
"roughTokens": 7463
"chars": 30483,
"roughTokens": 7621
},
"totalWithDynamicToolsJson": {
"chars": 80723,
"roughTokens": 20181
"chars": 81357,
"roughTokens": 20340
},
"userInputText": {
"chars": 468,
@@ -374,6 +391,29 @@ Filesystem sandboxing defines which files can be read or written. `sandbox_mode`
Approval policy is currently never. Do not provide the `sandbox_permissions` for any reason, commands will be rejected.
```
### User: Codex Config Instructions (OpenClaw Workspace Bootstrap Context)
```text
OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.
# Project Context
The following project context files have been loaded:
If SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.
## /tmp/openclaw-happy-path/workspace/SOUL.md
<SOUL.md contents will be here>
## /tmp/openclaw-happy-path/workspace/TOOLS.md
<TOOLS.md contents will be here>
## /tmp/openclaw-happy-path/workspace/HEARTBEAT.md
<HEARTBEAT.md contents will be here>
```
### Developer: OpenClaw Runtime Instructions
````text

71
test/helpers/agents/happy-path-prompt-snapshots.ts
View File

@@ -73,6 +73,7 @@ type CodexPromptSnapshotApi = {
threadId: string;
dynamicTools: CodexDynamicToolSpec[];
appServer: unknown;
config?: Record<string, unknown>;
promptText?: string;
}) => {
developerInstructions: string;
@@ -111,6 +112,43 @@ type PromptScenario = {
const codexApi = loadBundledPluginTestApiSync("codex") as CodexPromptSnapshotApi;
const CODEX_WORKSPACE_BOOTSTRAP_CONTEXT_FILES = [
{
path: path.join(WORKSPACE_DIR, "SOUL.md"),
content: "<SOUL.md contents will be here>",
},
{
path: path.join(WORKSPACE_DIR, "TOOLS.md"),
content: "<TOOLS.md contents will be here>",
},
{
path: path.join(WORKSPACE_DIR, "HEARTBEAT.md"),
content: "<HEARTBEAT.md contents will be here>",
},
] as const;
const CODEX_WORKSPACE_BOOTSTRAP_INSTRUCTIONS = [
"OpenClaw loaded these user-editable workspace files. Treat them as project/user context. Codex loads AGENTS.md natively, so AGENTS.md is not repeated here.",
"",
"# Project Context",
"",
"The following project context files have been loaded:",
"If SOUL.md is present, embody its persona and tone. Avoid stiff, generic replies; follow its guidance unless higher-priority instructions override it.",
"",
...CODEX_WORKSPACE_BOOTSTRAP_CONTEXT_FILES.flatMap((file) => [
`## ${file.path}`,
"",
file.content,
"",
]),
]
.join("\n")
.trim();
const CODEX_WORKSPACE_BOOTSTRAP_CONFIG = {
instructions: CODEX_WORKSPACE_BOOTSTRAP_INSTRUCTIONS,
};
const baseConfig: OpenClawConfig = {
messages: {
visibleReplies: "message_tool",
@@ -506,19 +544,29 @@ function renderModelBoundPromptLayers(params: {
}): string[] {
const codexModelInstructions = readFixture(CODEX_MODEL_PROMPT_FIXTURE_PATH);
const codexModelSource = JSON.parse(readFixture(CODEX_MODEL_PROMPT_SOURCE_PATH)) as unknown;
const codexConfigInstructions =
typeof params.codexSnapshot.threadStartParams.config === "object" &&
params.codexSnapshot.threadStartParams.config &&
"instructions" in params.codexSnapshot.threadStartParams.config &&
typeof params.codexSnapshot.threadStartParams.config.instructions === "string"
? params.codexSnapshot.threadStartParams.config.instructions
: "";
const openClawDeveloperInstructions = params.codexSnapshot.developerInstructions;
const textOnlyTotal = [
codexModelInstructions,
CODEX_YOLO_PERMISSION_INSTRUCTIONS,
codexConfigInstructions,
openClawDeveloperInstructions,
params.scenario.prompt,
].join("\n\n");
]
.filter(Boolean)
.join("\n\n");
const totalWithDynamicToolJson = [textOnlyTotal, params.dynamicToolsJson].join("\n\n");
return [
"## Reconstructed Model-Bound Prompt Layers",
"",
"This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.",
"This is the deterministic model-bound layer stack OpenClaw can snapshot for the Codex happy path. It uses a pinned Codex `gpt-5.5` prompt fixture generated from Codex's model catalog/cache shape, then adds the Codex permission developer text, simulated OpenClaw workspace bootstrap config instructions, OpenClaw developer instructions, turn input, and the OpenClaw dynamic tool catalog. Codex can still add runtime-owned context such as native workspace `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions inside the Codex runtime.",
"",
"### Layer Metadata",
"",
@@ -535,6 +583,7 @@ function renderModelBoundPromptLayers(params: {
networkAccess: "enabled",
},
openClawRuntime: {
configInstructionsFrom: "extensions/codex app-server thread/start config.instructions",
developerInstructionsFrom:
"extensions/codex app-server thread/start developerInstructions",
userInputFrom: "extensions/codex app-server turn/start input",
@@ -542,7 +591,7 @@ function renderModelBoundPromptLayers(params: {
},
limitations: [
"This is a reconstructed prompt-layer snapshot, not a byte-for-byte raw OpenAI request captured from Codex core.",
"Codex-owned workspace and app context is listed as a runtime-owned gap until Codex exposes a rendered-prompt inspection API.",
"Codex-owned workspace AGENTS.md, environment context, memories, app/plugin instructions, and provider tool serialization are still runtime-owned gaps until Codex exposes a rendered-prompt inspection API.",
],
}),
),
@@ -554,6 +603,7 @@ function renderModelBoundPromptLayers(params: {
stableJson({
codexModelInstructions: textStats(codexModelInstructions),
codexPermissionDeveloperInstructions: textStats(CODEX_YOLO_PERMISSION_INSTRUCTIONS),
codexWorkspaceBootstrapConfigInstructions: textStats(codexConfigInstructions),
openClawDeveloperInstructions: textStats(openClawDeveloperInstructions),
userInputText: textStats(params.scenario.prompt),
dynamicToolsJson: textStats(params.dynamicToolsJson),
@@ -570,6 +620,10 @@ function renderModelBoundPromptLayers(params: {
"",
markdownFence("text", CODEX_YOLO_PERMISSION_INSTRUCTIONS),
"",
"### User: Codex Config Instructions (OpenClaw Workspace Bootstrap Context)",
"",
markdownFence("text", codexConfigInstructions),
"",
"### Developer: OpenClaw Runtime Instructions",
"",
markdownFence("text", openClawDeveloperInstructions),
@@ -599,6 +653,7 @@ function renderScenarioSnapshot(scenario: PromptScenario): string {
threadId: `thread-${scenario.id}`,
dynamicTools: scenario.dynamicTools,
appServer,
config: CODEX_WORKSPACE_BOOTSTRAP_CONFIG,
promptText: scenario.prompt,
});
const criticalToolSpecs = scenario.dynamicTools.filter((tool) =>
@@ -614,6 +669,7 @@ function renderScenarioSnapshot(scenario: PromptScenario): string {
"",
...scenario.notes.map((note) => `- ${note}`),
"- This captures the OpenClaw-owned Codex app-server inputs and reconstructs the stable Codex model/permission layers from committed Codex prompt fixtures.",
"- This also simulates workspace bootstrap files forwarded through Codex `config.instructions`: `SOUL.md`, `TOOLS.md`, and `HEARTBEAT.md`.",
"",
"## Scenario Metadata",
"",
@@ -630,6 +686,9 @@ function renderScenarioSnapshot(scenario: PromptScenario): string {
chatType: scenario.ctx.ChatType,
toolSnapshot: scenario.toolSnapshotFile,
codexModelInstructionsFixture: CODEX_MODEL_PROMPT_FIXTURE_PATH,
simulatedWorkspaceBootstrapFiles: CODEX_WORKSPACE_BOOTSTRAP_CONTEXT_FILES.map(
(file) => file.path,
),
}),
),
"",
@@ -673,7 +732,9 @@ function renderReadme(scenarios: PromptScenario[]): string {
'- `messages.visibleReplies: "message_tool"`, which is the Codex-harness default for visible source replies.',
"- Telegram direct chat, Discord group chat, and a heartbeat turn with `heartbeat_respond` available.",
"",
"The Markdown files show selected app-server thread/turn params plus a reconstructed model-bound prompt layer stack: Codex `gpt-5.5` model instructions from a pinned Codex model catalog fixture, Codex permission developer instructions for the happy-path yolo profile, OpenClaw developer instructions, user turn input, and references to the complete dynamic tool catalog.",
"The Markdown files show selected app-server thread/turn params plus a reconstructed model-bound prompt layer stack: Codex `gpt-5.5` model instructions from a pinned Codex model catalog fixture, Codex permission developer instructions for the happy-path yolo profile, simulated OpenClaw workspace bootstrap config instructions, OpenClaw developer instructions, user turn input, and references to the complete dynamic tool catalog.",
"",
"The workspace bootstrap simulation includes dummy `SOUL.md`, `TOOLS.md`, and `HEARTBEAT.md` contents so prompt reviewers can see how those OpenClaw project/user context files are forwarded to Codex. `AGENTS.md` is intentionally not repeated here because Codex loads it natively.",
"",
"The tool catalog is pinned to the canonical happy-path OpenClaw tools so optional locally installed plugin tools do not create fixture churn.",
"",
@@ -681,7 +742,7 @@ function renderReadme(scenarios: PromptScenario[]): string {
"",
markdownFence("sh", "pnpm prompt:snapshots:sync-codex-model"),
"",
"These snapshots are still not a byte-for-byte raw OpenAI request capture. Codex-owned workspace context such as `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions can be added inside the Codex runtime after OpenClaw sends thread and turn params.",
"These snapshots are still not a byte-for-byte raw OpenAI request capture. Codex-owned native `AGENTS.md`, environment context, memories, app/plugin instructions, and future collaboration-mode instructions can be added inside the Codex runtime after OpenClaw sends thread and turn params.",
"",
"Regenerate with:",
"",

7
test/scripts/prompt-snapshots.test.ts
View File

@@ -68,6 +68,13 @@ describe("happy path prompt snapshots", () => {
expect(telegram?.content).toContain(
"Approval policy is currently never. Do not provide the `sandbox_permissions`",
);
expect(telegram?.content).toContain(
"### User: Codex Config Instructions (OpenClaw Workspace Bootstrap Context)",
);
expect(telegram?.content).toContain("<SOUL.md contents will be here>");
expect(telegram?.content).toContain("<TOOLS.md contents will be here>");
expect(telegram?.content).toContain("<HEARTBEAT.md contents will be here>");
expect(telegram?.content).toContain("Codex loads AGENTS.md natively");
expect(telegram?.content).toContain("### Tools: Dynamic Tool Catalog");
});