Dark-Alex-17/coyote
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • bd7cad21e2 feat(TASK-037): wire supervisor-agent and graph hook seams Dark-Alex-17 2026-09-15 19:46:41 -06:00
  • 0c4d0d777c feat(TASK-034): hook dispatch engine in src/hooks.rs Dark-Alex-17 2026-09-15 19:01:36 -06:00
  • b020e1b75e fix(TASK-033): cover hooks env overrides with tests and debug-log malformed hook config Dark-Alex-17 2026-09-15 16:06:03 -06:00
  • 9693cdd4ac feat(TASK-033): add hook config schema across all scopes Dark-Alex-17 2026-09-15 15:48:14 -06:00
  • a0cc5f291e docs(TASK-044): review polish — visible-output wording, web-search wire pin note, truncation debug line, thinking-exclusion pin test Dark-Alex-17 2026-09-16 13:42:42 -06:00
  • ce20927713 fix(TASK-044): streaming response.incomplete parity — keep partial output, record usage Dark-Alex-17 2026-09-16 13:31:52 -06:00
  • 9e7c957a3d docs(TASK-044): document wire_api selection, defaults, and wire-aware patch migration Dark-Alex-17 2026-09-16 13:21:10 -06:00
  • 3b66101361 feat(TASK-044): surface the streaming response.incomplete event with its reason Dark-Alex-17 2026-09-16 13:21:10 -06:00
  • 0dc9dd91c3 chore(TASK-043): regenerate model catalog Dark-Alex-17 2026-09-16 12:51:49 -06:00
  • 9825ccd31e feat(TASK-043): Responses usage parsing, failure hardening, responses default for stock openai Dark-Alex-17 2026-09-16 12:51:44 -06:00
  • 688ababecc test(TASK-042): prove serialize leg for old-session blocks and YAML round-trip Dark-Alex-17 2026-09-16 12:24:30 -06:00
  • d4605794e8 feat(TASK-042): Responses reasoning-item replay, strict:false, cross-client portability Dark-Alex-17 2026-09-16 12:13:21 -06:00
  • 3a2736b667 test(TASK-041): pin silent-ignore of malformed model-patch responses sub-key Dark-Alex-17 2026-09-16 11:58:12 -06:00
  • 003d036e31 docs: restore the loop-tracker exemption comment Dark-Alex-17 2026-09-16 11:52:09 -06:00
  • e4dfbe906c feat: updated sisyphus and architect agents so that when reviewing a PR, they add review round summaries as new comments rather than updating the PR description Dark-Alex-17 2026-09-16 11:36:34 -06:00
  • 57c1a86195 feat(TASK-041): add WireApi enum and wire-aware Responses dispatch Dark-Alex-17 2026-09-16 11:29:31 -06:00
  • 8be9c8e152 fix: exempt agent__check_inbox and agent__task_list from loop Dark-Alex-17 2026-09-16 11:07:10 -06:00
  • fa7f96e9e8 Merge remote-tracking branch 'refs/remotes/origin/main' Dark-Alex-17 2026-09-16 10:18:21 -06:00
  • 5046aebca3 feat: support config-driven OAuth endpoints on the openai and gemini clients Dark-Alex-17 2026-09-16 10:17:05 -06:00
  • 1d6823ca0c chore: refresh generated model catalog [skip ci] github-actions[bot] 2026-09-16 12:33:39 +00:00
  • 9fb90be851 fix: added additional constraints around parallelization bugs in verification tasks Dark-Alex-17 2026-09-16 00:36:33 -06:00
  • 4cc5563ba8 feat: support config-driven OAuth endpoints on the claude client Dark-Alex-17 2026-09-15 23:51:51 -06:00
  • fcc0049ab2 feat: further increased budgets for adversary and code reviewer agents Dark-Alex-17 2026-09-15 23:09:59 -06:00
  • dca9b1553e test: bumped the adversary's holistic max iterations from 20 to 30 in the tests Dark-Alex-17 2026-09-15 21:41:13 -06:00
  • ea31236d29 Merge branch 'main' of github.com:Dark-Alex-17/coyote Dark-Alex-17 2026-09-15 20:06:16 -06:00
  • d16db663df feat: increased resource budgets for review suite Dark-Alex-17 2026-09-15 20:06:07 -06:00
  • 94d48fb76e chore: refresh generated model catalog [skip ci] github-actions[bot] 2026-09-16 00:52:03 +00:00
  • 6c50c60c22 feat: improve the code review suite to support larger PRs Dark-Alex-17 2026-09-15 15:14:25 -06:00
  • f31885ba07 test: fixed tests for the new finding-verifier agent changes Dark-Alex-17 2026-09-15 15:04:48 -06:00
  • fd26f43b84 feat: Improved the code-reviewer suite so it doesn't need to loop as frequently for failed reviews based on the finding-verifier (which may need many round trips to verify large PRs) Dark-Alex-17 2026-09-15 14:50:39 -06:00
  • e240be8bda feat: improved the adversary agent to look for facts and not criteria. Dark-Alex-17 2026-09-15 12:02:16 -06:00
  • 9e2f3c0f12 feat: auto continuation is unavailable in llm nodes; continuation driven loops should be in agent nodes only Dark-Alex-17 2026-09-15 11:42:39 -06:00
  • 8746e3e29a Merge branch 'main' of github.com:Dark-Alex-17/coyote Dark-Alex-17 2026-09-15 11:14:24 -06:00
  • 5b9852d80e fix: Require injection of user interaction functions for LLM nodes to be explicit Dark-Alex-17 2026-09-15 11:13:12 -06:00
  • 99d4b20ac3 chore: refresh generated model catalog [skip ci] github-actions[bot] 2026-09-15 12:32:39 +00:00
  • 17864c8864 chore: refresh generated model catalog [skip ci] github-actions[bot] 2026-09-15 01:14:26 +00:00
  • 2a04c4f5e1 chore: bump Cargo.toml and sandbox image to 0.10.3 github-actions[bot] 2026-09-15 00:00:52 +00:00
  • ca58176ef9 bump: version 0.10.2 → 0.10.3 [skip ci] github-actions[bot] 2026-09-15 00:00:48 +00:00
  • cfb83a629e feat: create a config alias for default_reasoning_level (to match some wording and make the config flexible) Dark-Alex-17 2026-09-14 17:47:01 -06:00
  • 8377f8287b feat: support custom resource definitions for openai-compatible Oauth customization Dark-Alex-17 2026-09-14 17:46:44 -06:00
  • ffb75cb168 feat: Support specifying the compression model to use when compressing a session for greater cost controls Dark-Alex-17 2026-09-14 15:49:44 -06:00
  • 58b0dbd812 feat: Added full cost tracking to Coyote and added it to the default right prompt Dark-Alex-17 2026-09-14 15:24:48 -06:00
  • bd5c0a7cdd chore: updated the models.yaml to incorporate cached read/write prices pulled from models.dev and openrouter Dark-Alex-17 2026-09-14 14:47:22 -06:00
  • d84dea21b6 feat: added support for input/output token cache cost tracking Dark-Alex-17 2026-09-14 14:47:01 -06:00
  • 755b25f9b2 fix: empty tool results can't cause errors in bedrock, claude, openai, or vertexai clients anymore Dark-Alex-17 2026-09-14 14:46:46 -06:00
  • 23f941b09f fix: proper support for reasoning/thinking in bedrock models Dark-Alex-17 2026-09-14 14:46:15 -06:00
  • 77da48a40f feat: Support persisting paused todo lists to sessions so users can exit and resume paused todo sessions; Also fixed a bug so todo list pausing is only supported when being used in a session since it typically requires multi-turn communication Dark-Alex-17 2026-09-14 13:57:08 -06:00
  • 5d59adadbc feat: Support token caching to reduce LLM costs in Anthropic models (not typically handled automatically by APIs) Dark-Alex-17 2026-09-14 13:56:13 -06:00
  • 9a6ffcb8ca Merge pull request #28 from Dark-Alex-17/feat/PLAN-review-retry Alex Clarke 2026-09-14 11:46:15 -06:00
  • 62e6fafaef Merge pull request #26 from Dark-Alex-17/feat/PLAN-agent-node-resilience Alex Clarke 2026-09-14 11:35:24 -06:00
  • beca133082 style: removed unneccessary path prefixes and organized imports Dark-Alex-17 2026-09-14 11:03:43 -06:00
  • c7ec4ed124 docs: restored formatting to agent example config file Dark-Alex-17 2026-09-14 11:03:24 -06:00
  • cee5cf0f36 test(rag): make TempDir names unique across parallel tests Dark-Alex-17 2026-09-14 05:52:54 -06:00
  • 85cb7b0fbf test: normalize CRLF in the gauntlet README contract pin (Windows CI) Dark-Alex-17 2026-09-14 05:37:37 -06:00
  • d89592fee6 test: normalize CRLF in the gauntlet script/README source pins (Windows CI) Dark-Alex-17 2026-09-14 05:18:52 -06:00
  • 342f39f30e ci: run pull_request checks for PRs based on any branch Dark-Alex-17 2026-09-14 04:56:42 -06:00
  • 4f0c7e4431 test(TASK-030): pin fallback-note placement; detect concrete task ids (with illustrative-example allowlist) instead of run-local literals Dark-Alex-17 2026-09-14 04:29:26 -06:00
  • fcf67b07a7 test(TASK-030): pin placement, ladder clause, findings-first rule, README sync; reconcile gauntlet README caller guidance Dark-Alex-17 2026-09-14 04:21:42 -06:00
  • 78e326039d feat(TASK-030): callers escalate on GAUNTLET_REVIEW_INCOMPLETE instead of findings-routing Dark-Alex-17 2026-09-14 04:12:56 -06:00
  • f18c436e2c fix(TASK-032): enforce incomplete⇒BLOCKED structurally; anchor adversary degraded detection Dark-Alex-17 2026-09-14 03:58:56 -06:00
  • 786e5542d4 feat(TASK-032): review-gauntlet incomplete-review machine line + ×N attempt details Dark-Alex-17 2026-09-14 02:59:58 -06:00
  • 18f5ada585 feat(TASK-029): review-gauntlet fault-only lane re-run loop Dark-Alex-17 2026-09-14 01:41:52 -06:00
  • b90e04f227 fix(TASK-028): pin example-doc wording; make not-found escalation hint match queue semantics Dark-Alex-17 2026-09-14 00:39:20 -06:00
  • 3339920377 fix(TASK-028): pin global-template exclusion, tighten timeout boundary test, reword stale hint Dark-Alex-17 2026-09-14 00:31:34 -06:00
  • 9a505f951a feat(TASK-028): escalation_timeout 0 waits indefinitely and is the default Dark-Alex-17 2026-09-14 00:22:28 -06:00
  • b87367bb1b style: Refactored some imports and LLM hints Dark-Alex-17 2026-09-14 10:50:24 -06:00
  • ad3a5c6d49 test: normalize CRLF in the run_checks docstring pin (Windows CI) Dark-Alex-17 2026-09-13 20:14:09 -06:00
  • 68e893d684 fix(adversary): red verification runs block CONFORMS; re-derive timeout envelope; robust runner Dark-Alex-17 2026-09-13 19:39:34 -06:00
  • 156f3193a9 fix(review-suite): detect nested finding-verifier faults; harden render/fault markers Dark-Alex-17 2026-09-13 19:38:52 -06:00
  • 4555af8299 fix(review-gauntlet): summary-line critical count, nested-fault hook, fail-closed lane fallback Dark-Alex-17 2026-09-13 19:38:12 -06:00
  • ea76331c49 fix(graph): typed ApiStatusError transient classification; warn on bare output_schema Dark-Alex-17 2026-09-13 19:37:28 -06:00
  • 84b5a72eac fix(review-suite): mark example verification commands as placeholders; pin variable description and prompt note Dark-Alex-17 2026-09-13 16:57:17 -06:00
  • 27101909e0 fix(graph): auto-merge only schema-declared keys from llm output Dark-Alex-17 2026-09-13 16:43:08 -06:00
  • 300c7aeff0 fix(review-suite): pass verification_commands as a declared variable, never LLM-extracted from prose Dark-Alex-17 2026-09-13 16:27:49 -06:00
  • 00c3b60c2f feat(graph): stream llm-node completions so node timeout governs long generations Dark-Alex-17 2026-09-13 15:26:41 -06:00
  • 22421a4e74 test(adversary): pin run_checks process-group kill and deadline-vs-timeout ordering Dark-Alex-17 2026-09-13 11:39:41 -06:00
  • fa5ee5f4fc test(graphs): lock retry counts, fallback wiring, and check_criterion's read-only tools Dark-Alex-17 2026-09-13 11:34:41 -06:00
  • ba3a904baf test(graph): prove per-attempt timeout budget and cover extraction-fault text Dark-Alex-17 2026-09-13 11:30:45 -06:00
  • 8e42a59b98 docs(adversary): fix mermaid node label and criterion_fault join edge Dark-Alex-17 2026-09-13 11:29:31 -06:00
  • 081fa2c9d4 test(suite): replace plan-label references with self-contained wording Dark-Alex-17 2026-09-13 11:29:31 -06:00
  • 48753eaa46 fix(agents): scrub GRAPH_STATE* from adversary verification commands; raise review-suite max_agent_depth Dark-Alex-17 2026-09-13 03:30:15 -06:00
  • 62ecb5b415 test(adversary): pin exact died-evidence prefix on the criterion_fault crash path Dark-Alex-17 2026-09-13 02:54:09 -06:00
  • 7ada3fb47c fix(adversary): narrow died-marker predicate, degraded banner for died criteria, hermetic script tests Dark-Alex-17 2026-09-13 02:44:56 -06:00
  • 1cf42fffa2 fix(adversary): fail closed on PIPELINE-FAULT evidence regardless of status; pin bail-chain and validator shape Dark-Alex-17 2026-09-13 02:26:12 -06:00
  • 66b30bb06a test(code-reviewer): pin synthesize max_attempts and fail-closed wiring Dark-Alex-17 2026-09-13 00:15:02 -06:00
  • 3f49db86b2 fix(adversary): fail closed per criterion when check_criterion dies Dark-Alex-17 2026-09-13 00:13:54 -06:00
  • 7dac4620a8 fix(engine): carry full error chain in llm node fault text Dark-Alex-17 2026-09-13 00:10:24 -06:00
  • 5b4462135f style: cargo fmt after doc-comment cleanup Dark-Alex-17 2026-09-12 22:56:16 -06:00
  • fdae98adff docs: sync agent READMEs + diagrams with resilience hardening Dark-Alex-17 2026-09-12 22:53:11 -06:00
  • f230c28bf4 docs: removed references to individual task files in the code documentation Dark-Alex-17 2026-09-12 22:24:10 -06:00
  • 13b8eb87a9 fix(TASK-020): skip POSIX bash script tests on Windows Dark-Alex-17 2026-09-12 19:07:45 -06:00
  • 0ce2654173 feat(TASK-023): deep-research hardening — fail-safe scripts, fault fallbacks, pipeline notes Dark-Alex-17 2026-09-12 18:46:02 -06:00
  • c98b6bc9cd test(TASK-021): pin task-queue guidance text in shipped artifacts Dark-Alex-17 2026-09-12 18:24:27 -06:00
  • f2d90ed7a6 docs(TASK-021): fix task-queue example IDs + document dispatch/collect/failure semantics Dark-Alex-17 2026-09-12 18:16:11 -06:00
  • 22976a79f9 fix(TASK-020): surface step-runner LLM-fault captures where the fallback lands Dark-Alex-17 2026-09-12 18:11:13 -06:00
  • 4263d684e9 feat(TASK-020): finding-verifier + step-runner fail-closed hardening Dark-Alex-17 2026-09-12 17:49:10 -06:00
  • a8496aa50e fix(TASK-019): carry PIPELINE-FAULT prefix through verdict.py's crash guard Dark-Alex-17 2026-09-12 17:31:41 -06:00
  • 4e72c6f5e6 feat(TASK-019): code-reviewer fail-closed review pipeline Dark-Alex-17 2026-09-12 17:16:48 -06:00
  • ebb9e9af62 fix(TASK-018): harden verdict_gate fault routing and widen default_lanes fallback Dark-Alex-17 2026-09-12 16:33:38 -06:00
  • 80c20afcdd test(TASK-018): cover parse_fault.py fault normalization Dark-Alex-17 2026-09-12 16:01:42 -06:00