S-QUORUM: agent tool_call_log + capability claiming filter (gated) + CI gate#25
Merged
Merged
Conversation
…ed) + CI gate S-QUORUM Phases 4, 5, 7 (trinity-symphony-shared). All flag-gated, default-off; both verified not to alter current behavior when flags are unset. Phase 4 - tool_call_log (gated TOOL_CALL_LOGGING): - lib/tool-call-logger.js: logToolCall() - sha256(output), jsonb input, inserts via direct-pg pooler into the hash-chained tool_call_log; never throws. - Wired into callLLM (fire-and-forget; no latency, no-op unless flag set). Logs tool_name=llm:<provider>:<model> + prompt_sha256. Write path + hash-chain trigger verified live (row inserted, previous_entry_hash stamped 64-char sha256, then removed). Phase 5 - capability claiming filter (gated CAPABILITY_FILTER, default OFF): - getNextTask query gains a predicate that, when capFilter is null (flag off), is a no-op (byte-identical to prior behavior). When on, agents only claim AGENT_TASK_TYPES (default = the 9 knownTypes), so cait/EVERGREEN (~84% of shadow_rejects, per S-HARDEN) stop being claimed-then-rejected. Phase 7 - .github/workflows/ci.yml: syntax-check core libs + run the node test suite. Node tests green (getNextTask 8/8, provenance, pulseCheckExecutor 9/9). Deconfliction: getNextTask/callLLM overlap GA's in-flight feat/ga-2026-05-30-t12-concurrency - these changes are small + flag-gated for easy folding into T12. On branch; no merge. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
This pull request has been ignored for the connected project Preview Branches by Supabase. |
…node tests green) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
S-QUORUM Phases 4, 5, 7 for the swarm agents. All flag-gated default-off, on-branch, no merge. Node tests green (getNextTask 8/8, provenance, pulseCheckExecutor 9/9).
Phase 4 — tool_call_log (gated TOOL_CALL_LOGGING)
lib/tool-call-logger.js:logToolCall()— sha256(output), jsonb input, inserts via the direct-pg pooler into the hash-chainedtool_call_log; never throws.callLLM(fire-and-forget — no latency, no-op unless flag set). Write path + hash-chain trigger verified live (row inserted,previous_entry_hashstamped, then removed).Phase 5 — capability claiming filter (gated CAPABILITY_FILTER, default OFF)
getNextTaskquery gains a capability predicate that is a no-op when the flag is off (byte-identical). When on, agents only claimAGENT_TASK_TYPES(default = the 9 knownTypes), socait/EVERGREEN(~84% of shadow_rejects per S-HARDEN) stop being claimed-then-rejected.Phase 7 — CI
.github/workflows/ci.yml: syntax-check core libs + run the node test suite.getNextTask/callLLMoverlap GA's in-flightfeat/ga-2026-05-30-t12-concurrency. These changes are intentionally small + flag-gated for easy folding into T12 — GA/Sean to fold, not blind-merge. No self-merge.🤖 Generated with Claude Code