8000 M11: autonomous by default; only one-way doors go to the human by adeelahmad · Pull Request #122 · adeelahmad/aloop · GitHub
[go: up one dir, main page]

Skip to content

M11: autonomous by default; only one-way doors go to the human - #122

Merged
adeelahmad merged 5 commits into
masterfrom
claude/project-thread-s19a0c
Nov 19, 2026
Merged

adeelahmad merged 5 commits into
masterfrom
claude/project-thread-s19a0c

Conversation

@adeelahmad
@adeelahmad adeelahmad commented Nov 19, 2026 •
Copy link
Copy Markdown
Owner

Requested by Adeel · project thread

What

Core rule (owner, 2026-11-19): generation is cheap, so only one-way-door decisions go to the human. ROADMAP §4.14 has the audit (A1–A12, every place a run waited for the human, with file:line) and the design.

Implemented:

  • M11.1 Autonomous is the default mode (events authority without --mode); reviewed is opt-in (AGENTIC_AUTONOMY_MODE=reviewed).
  • M11.2 The typed intent is the grant: a sprint planned from an intent whose words are exactly a captured human prompt (≥4 words, optionally after /aladdin) gets autonomy with no autonomous SN <code> line. An untyped paraphrase, a short reply, or a plan re-pointed at another intent still asks / loses the grant.
  • M11.4 An intake ask with the intake's recommended answer and no door is taken and recorded (intent.normalized taken=[…]); one naming a one-way door, or with no recommendation, still asks.
  • M11.10 New door-classifier role (planning route) tags each planned task one-way (mission doors tag T --door ID) or two-way. mission doors plan SN lists the open one-way doors together; gate-door-plan (SubagentStop) blocks the classifier's stop while a task is untagged or a wall is planned; gate-dispatch parks only tasks with an unauthorized door. Tags only add, and the PreToolUse door gate stays the backstop.
  • M11.11 Override: ALOOP_DOORS_TWO_WAY=<door,…> makes a door two-way only while a door.reclassified ledger row holds. That row counts only with the human's typed two-way door:<id> <code> line (M9.17 provenance), so an agent can't forge it. Walls and question-only doors are refused, and revoke closes it.
  • M11.6 Init asks nothing by default (--ask brings the questions back).
  • Docs aligned: OVERVIEW, README, plugin README, SKILL, init skill, intake persona, CLAUDE.md, site, ROADMAP §0/§4.14/§7, TRACKING.

No gate's deterministic checks were weakened. Every grant still rests on a captured human prompt.

Call: an untagged task is treated as two-way at dispatch (the door gate is the backstop), so older plans and fixtures keep running. The classifier's own stop gate is what requires every tag.

Next: M11.5 sprint chaining, M11.3 plan judge, M11.7, M11.8, M11.9 live run.

Evidence

  • New checks were red on master before the code changed, green after: plan-approval.test.py 22, intent-contract.test.py 16.
  • New suites: mission-door-plan.test.py 6, mission-doors-reclassify.test.py 5.
  • make pycheck test-hosts green, make eval-validate (18 suites), check-tracking (111 requirements).

🤖 Generated with Claude Code

https://claude.ai/code/session_01FeZyj9c1mTZ8GsF1SyvzYR

@adeelahmad adeelahmad self-assigned this Nov 19, 2026
…rant; a two-way ask takes its recommended answer

Only one-way doors go to the human (generation is cheap: a reversible choice is redone).
- events authority: --mode defaults to autonomous (AGENTIC_AUTONOMY_MODE=reviewed opts in)
- _approval: a sprint planned from an intent whose words are exactly a captured human prompt is
  granted autonomy with no second line; untyped words, a short reply or a re-pointed plan still ask
- intent: in autonomous mode an ask with a recommended answer and no door is taken and recorded
  (intent.normalized taken=[...]); a door or no recommendation still asks
Tests: plan-approval 22 (new check red on master), intent-contract 16 (new check red on master).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FeZyj9c1mTZ8GsF1SyvzYR
…-question init; docs aligned

Only one-way doors go to the human; the door-classifier decides which, before execution.
- agents/door-classifier.md (planning route) + gate-door-plan (SubagentStop): every task tagged
- mission doors tag/plan: tags only add; open one-way doors listed together; gate-dispatch
  parks only those tasks
- mission doors reclassify: ALOOP_DOORS_TWO_WAY plus the human's typed 'two-way door:<id> <code>'
  (ledger row door.reclassified, M9.17 provenance); walls and question-only doors refused
- init skill applies the defaults with no questions (--ask restores them)
- OVERVIEW, README, plugin README, SKILL, CLAUDE.md, site, ROADMAP/TRACKING aligned
Tests: mission-door-plan 6, mission-doors-reclassify 5, make test-hosts green, eval-validate,
check-tracking.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FeZyj9c1mTZ8GsF1SyvzYR
@adeelahmad adeelahmad changed the title ROADMAP M11: autonomous by default (audit of every human stop + design) M11: autonomous by default; only one-way doors go to the human Nov 19, 2026
@adeelahmad
adeelahmad marked this pull request as ready for review November 19, 2026 13:35
… a typed prompt can be the grant

- _approval.said: the whole prompt or its first line (after an optional slash command)
- e2e-live.sh: no approval line; the prompt's first line is the intent
- intake persona: --raw is the human's words verbatim
Tests: plan-approval 22, e2e-live-hosts 16.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FeZyj9c1mTZ8GsF1SyvzYR
…rant, 12/12, $0.82

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FeZyj9c1mTZ8GsF1SyvzYR
@adeelahmad
adeelahmad merged commit 6d339a1 into master Nov 19, 2026
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants

0