zero/internal at perf/deterministic-completion-policy - zero - Free-Git: For Devs Not Bully

Twigpine/zero

0 Fork 0

mirror of https://github.com/Twigpine/zero.git synced 2026-11-11 23:40:28 -10:00

Files

T

History

anandh8x a3b09218aa Add deterministic completion policy

Extract completion decisions from the agent loop into a feature-gated typed policy with complete, incomplete, and uncertain outcomes. Preserve bounded plan-stall nudges and allow at most one semantic acceptance check for self-correcting runs.

Tested: make build

Tested: make test

Tested: go fmt ./...

Tested: go vet ./...

Tested: make lint

Tested: govulncheck ./...

Note: repository-wide pinned golangci-lint reports 36 pre-existing unrelated findings; internal/agent/... reports 0 issues.

2026-07-18 00:21:22 +05:30

..

fix(acp): make truncateHint rune-safe (#614 )

2026-07-10 10:05:56 +08:00

Add deterministic completion policy

2026-07-18 00:21:22 +05:30

test: widen timing margins on two Windows-flaky CI tests (#359 )

2026-07-01 21:28:52 +05:30

feat: five capability features from the reference-agent comparison (#276 )

2026-06-20 23:12:15 +05:30

feat(aimlapi): AI/ML API provider with guided onboarding (top-up + key issuance) (#655 )

2026-07-17 12:49:38 +08:00

fix(windows): resolve absolute path for taskkill to prevent hijacking (#617 )

2026-07-10 10:08:11 +08:00

OAuth / account login for providers (OpenRouter, xAI) + subscription-proxy support (#217 )

2026-06-16 15:33:36 +05:30

perf(output): add token-aware semantic output budgeting (PR11) (#717 )

2026-07-18 00:02:43 +05:30

fix(config): enforce MCP trust boundary so project config cannot override user disable (#609 )

2026-07-17 13:03:32 +08:00

fix(sandbox): scrub sensitive credentials from sandbox environment (#660 )

2026-07-14 21:46:10 +08:00

fix(securefile,credstore): call Sync on temp file before close and rename (#631 )

2026-07-12 22:35:55 +08:00

fix(sandbox): scrub sensitive credentials from sandbox environment (#660 )

2026-07-14 21:46:10 +08:00

fix(lock): prevent POSIX lock file overwrite and leak on Windows/Unix (#628 )

2026-07-13 21:45:08 +08:00

feat: voice dictation (speech-to-text) (#557 )

2026-07-07 11:13:14 +08:00

feat(providers): OAuth login profiles and list-first /provider manager (#560 )

2026-07-07 10:34:59 +08:00

fix(auth): persist OpenRouter API key after login (#595 )

2026-07-08 23:04:57 +08:00

fix(sandbox): scrub sensitive credentials from sandbox environment (#660 )

2026-07-14 21:46:10 +08:00

fix(lock): prevent POSIX lock file overwrite and leak on Windows/Unix (#628 )

2026-07-13 21:45:08 +08:00

fix(tools): Windows cmd.exe quoting guidance and clipboard escaping fix (#468 )

2026-07-09 09:23:40 +08:00

Add local control browser and terminal tools (#319 )

2026-06-26 08:22:56 +05:30

fix(keyring): pass generic password via stdin on macOS (#574 )

2026-07-14 14:46:09 +08:00

Add local control browser and terminal tools (#319 )

2026-06-26 08:22:56 +05:30

fix(lock): prevent POSIX lock file overwrite and leak on Windows/Unix (#628 )

2026-07-13 21:45:08 +08:00

feat: agent quality, caching, retry, and tooling upgrades (#506 )

2026-07-05 17:25:26 +08:00

fix: harden MCP credential boundaries (#597 )

2026-07-10 09:00:17 +08:00

token efficiency: lossless cost accounting + minified file reads (read_minified_file) (#256 )

2026-06-19 12:03:07 +05:30

feat(providers): refresh MiniMax model coverage (#665 )

2026-07-13 22:33:55 +08:00

Align sandbox permission blocks and backend states (#253 )

2026-06-19 19:50:12 +05:30

feat(npm): ship the native binary as platform optionalDependencies (#626 )

2026-07-11 23:25:41 +08:00

fix(lock): prevent POSIX lock file overwrite and leak on Windows/Unix (#628 )

2026-07-13 21:45:08 +08:00

Real-world hardening: graceful shutdown + crash capture (#136 )

2026-06-08 08:55:36 +05:30

perf(turn-bench): Phase 0 — strengthen oracles so pass rate can't be misread as correctness (#712 )

2026-07-17 17:32:45 +05:30

feat(skills): discover shared ~/.agents/skills with multi-root skill loading (#696 )

2026-07-16 21:59:20 +08:00

providercatalog

feat(aimlapi): AI/ML API provider with guided onboarding (top-up + key issuance) (#655 )

2026-07-17 12:49:38 +08:00

fix: allow non-TLS connections to private-network provider endpoints (#444 )

2026-07-04 08:35:14 +08:00

providermodelcatalog

feat(aimlapi): AI/ML API provider with guided onboarding (top-up + key issuance) (#655 )

2026-07-17 12:49:38 +08:00

providermodeldiscovery

feat(providers): add AI/ML API preset (rebased onto main) (#621 )

2026-07-16 22:57:35 +08:00

fix: resolve 15 findings from the 2026-06-20 deep audit (#281 )

2026-06-21 00:44:23 +05:30

provideronboarding

wedge stage 13: launch readiness (first-run + benchmark) (#186 )

2026-06-13 11:10:37 +05:30

feat(providers): add AI/ML API preset (rebased onto main) (#621 )

2026-07-16 22:57:35 +08:00

feat(reasoning): typed per-provider reasoning capability from a models.dev catalog (#338 )

2026-06-28 14:15:59 +05:30

fix model-discovery fallback + redact opaque Authorization tokens (#266 )

2026-06-20 10:39:35 +05:30

fix(update): clearer error on unsupported release platform (android/termux) (#603 )

2026-07-09 09:34:48 +08:00

feat(tui): clickable plan-step detail + plan/sidebar UX & rendering polish (#315 )

2026-06-25 20:23:29 +05:30

feat: add zero changes push and pr subcommands, and extra repo-info metrics (#391 )

2026-07-05 17:03:29 +08:00

Add shared workspace index foundation (#159 )

2026-06-10 17:30:55 +05:30

Remove Bun installer test dependency (#91 )

2026-06-06 13:58:08 +05:30

fix(sandbox): scrub dynamic credential env vars (#682 )

2026-07-17 13:04:12 +08:00

TUI scrollback re-architecture, Lime chat surface, and deep-audit fixes (#158 )

2026-06-10 17:18:30 +05:30

redaction: catch compound secret keys + non-bearer auth schemes; stop scanner over-redaction (#249 )

2026-06-18 20:01:25 +05:30

fix(securefile,credstore): call Sync on temp file before close and rename (#631 )

2026-07-12 22:35:55 +08:00

Add runtime permission and verification event contracts (#82 )

2026-06-06 13:13:08 +05:30

fix(sandbox): scrub sensitive credentials from sandbox environment (#660 )

2026-07-14 21:46:10 +08:00

feat(skills): discover shared ~/.agents/skills with multi-root skill loading (#696 )

2026-07-16 21:59:20 +08:00

fix(specialist): cap max specialist nesting depth (#491 )

2026-07-05 11:38:15 +08:00

Add spec mode review flow (#145 )

2026-06-08 21:07:57 +05:30

fix(openai): handle Ollama reasoning stream deltas (#486 )

2026-07-05 17:05:54 +08:00

fix(swarm): wait for job.Runs directly in scheduler skip test (#667 )

2026-07-15 21:05:46 +08:00

Move build smoke tooling to Go (#86 )

2026-06-06 12:25:25 +05:30

perf(output): add token-aware semantic output budgeting (PR11) (#717 )

2026-07-18 00:02:43 +05:30

perf(output): add token-aware semantic output budgeting (PR11) (#717 )

2026-07-18 00:02:43 +05:30

feat(aimlapi): AI/ML API provider with guided onboarding (top-up + key issuance) (#655 )

2026-07-17 12:49:38 +08:00

fix(update): clearer error on unsupported release platform (android/termux) (#603 )

2026-07-09 09:34:48 +08:00

polish: cut per-turn overhead, stabilize cache, add /new (#478 )

2026-07-04 20:56:52 +08:00

feat: five capability features from the reference-agent comparison (#276 )

2026-06-20 23:12:15 +05:30

Add runtime permission and verification event contracts (#82 )

2026-06-06 13:13:08 +05:30

Add shared workspace index foundation (#159 )

2026-06-10 17:30:55 +05:30

Clean up public release docs (#339 )

2026-06-28 17:13:21 +05:30

feat: gate project-scoped hooks, plugins, and MCP servers behind workspace trust (#529 )

2026-07-09 10:51:09 +08:00

Deep-audit remediation + sandbox hardening (batches 1–35) (#189 )

2026-06-13 21:32:48 +05:30

feat(providers): OAuth login profiles and list-first /provider manager (#560 )

2026-07-07 10:34:59 +08:00

feat: add zero changes push and pr subcommands, and extra repo-info metrics (#391 )

2026-07-05 17:03:29 +08:00

feat(openai): forward prompt_cache_key for server-side prefix cache routing (#515 )

2026-07-05 21:41:03 +08:00