Adds 5 vitest cases for postSlackMessageBestEffort's silent retry
behavior when Slack rejects a chat:write.customize-identity post:
- Retry on err.data.needed matching chat:write.customize
- Retry on chat:write.customize in response_metadata.acceptedScopes
- Retry on chat:write.customize in response_metadata.scopes
- Rethrow on different missing_scope (e.g. channels:history)
- Rethrow when identity is empty (hasCustomIdentity returns false)
Honor explicit strict-agentic execution contracts for incomplete-turn retry guards across providers, including local/compatible models that opt in without relying on OpenAI model inference.
Validation:
- pnpm test src/agents/pi-embedded-runner/run.incomplete-turn.test.ts
- pnpm check:changed
- GitHub CI + parity gate green
Thanks @ziomancer.
Normalize Anthropic-owned Opus 4.7 context reporting to 1M while keeping inferred and bare discovery paths conservative.
- normalize Anthropic and claude-cli Opus 4.7 runtime/status context metadata to 1M
- keep inferred-provider and bare discovery ids on discovered conservative limits
- add regression coverage for provider, lookup, status, and discovery-cache paths
- keep the Telegram abort-signal wrapper typing narrow so changed-scope validation stays green
Restore the Pi embedded session tool allowlist for OpenAI/OpenAI Codex GPT-5 runs and compaction sessions after Pi 0.68.1 began treating session tools as a global allowlist.
Local validation: pnpm check:changed.
GitHub validation: check/check-additional/node shards green; parity gate red on unrelated config.patch stale/rate-limit QA harness scenario after plugins.allow restart.
Verify Claude CLI session transcripts before reuse and clear phantom bindings with transcript-missing instead of passing stale --resume ids.\n\nFixes #70177.
* fix(amazon-bedrock): inject cache points for application inference profile ARNs
pi-ai's internal supportsPromptCaching checks model.id for specific Claude
model name patterns (e.g. "-4-", "claude-3-7-sonnet"), which fails for
application inference profile ARNs that don't contain the model name.
This causes prompt caching to silently break for Bedrock users with
application inference profiles.
Work around this by detecting when pi-ai would miss cache point injection
(via piAiWouldInjectCachePoints mirror) and patching the Converse API
payload via onPayload to add cachePoint blocks to the system prompt and
last user message — matching the same format pi-ai uses natively.
The fix is safe:
- Checks for existing cache points to avoid double-injection
- Respects cacheRetention: "none"
- Defaults to "short" retention (matching pi-ai default)
- Becomes a no-op once upstream pi-mono#2925 is fixed
Fixes#19279
Upstream: https://github.com/badlogic/pi-mono/issues/2925
* fix(amazon-bedrock): tighten app-profile cache injection
---------
Co-authored-by: Your Name <you@example.com>
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
#16763 added `onTimeout: "return"` with `timeoutMilliseconds: 10_000`
(grammY default). In practice, Telegram's webhook servers abort the
read well before 10s when handler latency is LLM-bound: `getWebhookInfo`
reports `last_error_message: "Read timeout expired"` and pending updates
pile up, cascading into multi-minute reply lag.
Reproducible A/B on identical infra (same region, same bot token):
- Minimal Python echo bot: 5 back-to-back webhook RTTs 341-642ms, clean.
- OpenClaw current main: intermittent Read timeout expired, 1-5 min lag.
The handler still runs to completion; only the Telegram-facing ack is
sooner. grammY's deployment guide suggests 5s for long-running handlers.
No new config surface; minimal one-line change to the existing constant
and its test assertion. If a configurable timeout is wanted, that can be
a follow-up (see stale #7754).
