Repository navigation
Description
Preset ID
autonomous-run-governance
Preset Name
Autonomous Run Governance
Version
0.4.12
Description
Adds permission-bounded autonomous delivery and an optional policy-driven intake-review gate before feature creation.
Author
Thorsten Hindermann
Repository URL
https://github.com/hindermath/spec-kit-preset-autonomous-run-governance
Download URL
Documentation URL
https://github.com/hindermath/spec-kit-preset-autonomous-run-governance/blob/v0.4.12/README.md
License
MIT
Required Spec Kit Version
=0.8.3
Required Extensions (optional)
Templates Provided
- autonomous-run-governance-model-routing — model-routing.json
- constitution-template — templates/constitution-addendum.md
- spec-template — templates/spec-addendum.md
- plan-template — templates/plan-addendum.md
- tasks-template — templates/tasks-addendum.md
- autonomous-run-agent-guidance-addendum-template — templates/agent-file-addendum.md
- autonomous-runbook-template — templates/autonomous-runbook-template.md
- autonomous-run-evidence-template — templates/autonomous-run-evidence-template.md
- autonomous-run-state-template — templates/autonomous-run-state-template.json
- autonomous-runner-profiles-template — templates/autonomous-runner-profiles-template.json
- autonomous-run-gate-requirements-template — templates/autonomous-run-gate-requirements-template.json
- autonomous-run-gate-evidence-template — templates/autonomous-run-gate-evidence-template.json
- autonomous-phase-result-template — templates/autonomous-phase-result-template.json
- autonomous-run-retrospective-template — templates/autonomous-run-retrospective-template.md
- autonomous-run-readiness-checklist-template — templates/autonomous-run-readiness-checklist-template.md
- autonomous-run-field-validation-summary — templates/field-validation-summary.md
Commands Provided
- speckit.autonomous — Execute a complete Spec Kit feature under explicit delivery authority
- speckit.autonomous-retrospective — Extract and classify reusable learning from a completed autonomous run
- speckit.autonomous-status — Inspect an autonomous run without changing repository state
- speckit.autonomous-stop — Gracefully pause an autonomous run at a safe boundary
- speckit.autonomous-resume — Resume a paused or interrupted autonomous run after audit
Number of Scripts (optional)
11
Tags
autonomous, governance, evidence, permissions, checkpoint
Key Features
- This existing submission is updated from v0.4.11 to the latest stable v0.4.12, published November 13, 2026: https://github.com/hindermath/spec-kit-preset-autonomous-run-governance/releases/tag/v0.4.12 . Only Autonomous Run Governance is submitted; the current catalog baseline is v0.4.4.
- Permission-bounded autonomous Spec Kit delivery with explicit lifecycle/stop/resume authority, semantic phase results, exact intended/index delivery sets and separate PreMerge/PostMerge evidence. Sixteen template contributions, five commands and eleven runtime scripts/helpers; priority 70; Spec Kit >=0.8.3; no required Spec Kit extensions.
- Security baseline versus script changes: v0.4.12 packages the bilingual SECURITY.md, confidential reporting route and explicit threat model: the checked worktree stays unchanged during validation, while its contents remain untrusted. CodeQL analyses are available on release commit
7fd8c95f567f5c957cedf5d3bf803b1c405ec42c. Automated checks cover their configured scope and do not certify finding-free code. Policy: https://github.com/hindermath/spec-kit-preset-autonomous-run-governance/blob/v0.4.12/SECURITY.md . - Scoped CodeQL investigation: the owner-confirmed investigation covered 15 alerts at nine distinct locations across four script-heavy presets (Autonomous Run, Intake Authoring, Intake Review and Intake Sequencing). They were false positives under the agreed stable-worktree threat model, so those alerts required no script-code corrections. This aggregate is not a claim that all 15 alerts belong to this preset or that all fourteen presets received corrected scripts.
- Exact release delta: every runtime script, command, template and routing file is byte-identical between v0.4.11 and v0.4.12. This is a security-policy/documentation patch. Earlier runtime changes, including the NUL-safe Git inventory correction in v0.4.10 and the separately recorded v0.4.11 changes, remain historical changes. Version-bound release evidence: https://github.com/hindermath/spec-kit-preset-autonomous-run-governance/blob/v0.4.12/docs/release-0.4.12-validation.md .
- Separate infrastructure hardening: physical repository/fleet path boundaries against parent symlinks were corrected in central Level 0 and distributed under a separate assignment. Those central script changes are not attributed to this preset patch. The cross-preset security basis (policy, CodeQL, private reporting, version-bound releases and separate source/installed locks) and central hardening are documented at https://github.com/hindermath/home-baseline/blob/main/docs/maintenance/preset-security-patch-rollout-2026-11-13.md . Mentioning that evidence neither submits other presets nor starts another rollout.
- ZIP SHA-256:
3b6282409d323e6f87fcc6f5c5923064bbdc52117b253bf909b61f4881034f81. All 86 archive files (456897 bytes uncompressed) were compared byte-for-byte with release/tag commit7fd8c95f567f5c957cedf5d3bf803b1c405ec42c. - Fresh isolated installation from the exact submitted URL succeeds at priority 70. All sixteen templates and five commands resolve through Spec Kit's type-aware manifest layer stack. The current official community-submission validator passes for the archive, README and submitted fields.
- Source PR feat(templates): integrate DSPy GEPA optimization + tooling #36 passed twelve checks, including reader-path, evidence-integrity and documentation checks on macOS/Ubuntu/Windows plus CodeQL checks: release: package security baseline as v0.4.12 hindermath/spec-kit-preset-autonomous-run-governance#36 . Fresh local documentation/release checks pass 126 documentation assertions, six release-documentation assertions and 61 release-status/history assertions. Fresh Bash-first cross-shell evidence-integrity regressions also pass all seven native Git-filename cases, intended/index delivery-set checks, semantic phase-result/lifecycle fixtures and model-routing negatives. Installation and validation start no autonomous product run or model invocation.
- Practical use and continuous improvement: The maintainer uses the preset family, including Autonomous Run Governance, in public Level-0/Level-2 repositories and private repositories. Public examples include home-baseline, TinyCalc, Show-CommandTui400 and absdd-image-sandbox. Feedback from ongoing development and delivery informs improvements to scripts, regression tests, workflow contracts and documentation. Practical experience complements version-bound testing and scoped security review; it does not establish security certification or universal acceptance of a release.
- Existing real-project v0.4.12 installation and matrix evidence is recorded in the linked November 13 Level-0 rollout record. A fresh read-only check confirms the existing Level-0 installed manifest matches the submitted release; no installation there is performed for this issue update. Historical product field runs are documented separately in the package field-validation summary and are not claimed as new v0.4.12 product or accessibility acceptance.
Testing Checklist
- Preset installs successfully via
specify preset add - Template resolution works correctly after installation
- Documentation is complete and accurate
- Tested on at least one real project
Submission Requirements
- Valid
preset.ymlmanifest included - Linked README (Documentation URL) explains how to use this preset and includes a valid
specify preset add ...command (preferablyspecify preset add --from <download-url>using the exact download URL) - LICENSE file included
- GitHub release created with version tag
- Preset ID follows naming conventions (lowercase-with-hyphens)
AI disclosure: Updated on behalf of @hindermath by OpenAI Codex (GPT-6.1; autonomous execution of the explicitly authorized single-preset issue update). Text and checks are AI-generated or automated. The scoped CodeQL triage interpretation was supplied/confirmed by the owner; archive and runtime-delta comparisons were independently repeated. No new human/product/accessibility approval is claimed. No merge, Home sync or rollout is authorized or performed by this update.
Activity
[-][Preset]: Update Autonomous Run Governance to v0.4.11[/-][+][Preset]: Update Autonomous Run Governance to v0.4.12[/+]github-actions commentedon Nov 16, 2026
cc @github/spec-kit-maintainers — new catalog submission for review.
Update Autonomous Run Governance preset to v0.4.12
github-actions commentedon Nov 16, 2026
Pull request created: #5446