loader/.cursor/skills/autodev/protocols.md

Autodev Protocols

User Interaction Protocol

Every time the autodev or a sub-skill needs a user decision, use the Choose A / B / C / D format. This applies to:

• State transitions where multiple valid next actions exist
• Sub-skill BLOCKING gates that require user judgment
• Any fork where the autodev cannot confidently pick the right path
• Trade-off decisions (tech choices, scope, risk acceptance)

When to Ask (MUST ask)

• The next action is ambiguous (e.g., "another research round or proceed?")
• The decision has irreversible consequences (e.g., architecture choices, skipping a step)
• The user's intent or preference cannot be inferred from existing artifacts
• A sub-skill's BLOCKING gate explicitly requires user confirmation
• Multiple valid approaches exist with meaningfully different trade-offs

When NOT to Ask (auto-transition)

• Only one logical next step exists (e.g., Problem complete → Research is the only option)
• The transition is deterministic from the state (e.g., Plan complete → Decompose)
• The decision is low-risk and reversible
• Existing artifacts or prior decisions already imply the answer

Choice Format

Always present decisions in this format:

══════════════════════════════════════
 DECISION REQUIRED: [brief context]
══════════════════════════════════════
 A) [Option A — short description]
 B) [Option B — short description]
 C) [Option C — short description, if applicable]
 D) [Option D — short description, if applicable]
══════════════════════════════════════
 Recommendation: [A/B/C/D] — [one-line reason]
══════════════════════════════════════

Rules:

1. Always provide 2–4 concrete options (never open-ended questions)
2. Always include a recommendation with a brief justification
3. Keep option descriptions to one line each
4. If only 2 options make sense, use A/B only — do not pad with filler options
5. Play the notification sound (per .cursor/rules/human-attention-sound.mdc) before presenting the choice
6. After the user picks, proceed immediately — no follow-up confirmation unless the choice was destructive

Optional Skill Gate (reusable template)

Several flow steps ask the user whether to run an optional skill (security audit, performance test, etc.) before auto-chaining. Instead of re-stating the Choose block and skip semantics at each such step, flow files invoke this shared template.

Template shape

══════════════════════════════════════
 DECISION REQUIRED: <question>
══════════════════════════════════════
 A) <option-a-label>
 B) <option-b-label>
══════════════════════════════════════
 Recommendation: <A|B> — <reason>
══════════════════════════════════════

Semantics (same for every invocation)

• On A → read and execute the target skill's SKILL.md; after it completes, auto-chain to <next-step>.
• On B → mark the current step skipped in the state file; auto-chain to <next-step>.
• On skill failure → standard Failure Handling (§Failure Handling) — retry ladder, then escalate via Choose block.
• Sound before the prompt — follow .cursor/rules/human-attention-sound.mdc.

How flow files invoke it

Each flow-file step that needs this gate supplies only the variable parts:

Action: Apply the **Optional Skill Gate** (protocols.md → "Optional Skill Gate") with:
- question:        <Choose-block header>
- option-a-label:  <one-line A description>
- option-b-label:  <one-line B description>
- recommendation:  <A|B> — <short reason, may be dynamic>
- target-skill:    <.cursor/skills/<name>/SKILL.md, plus any mode hint>
- next-step:       Step <N> (<name>)

The resolved Choose block (shape above) is then rendered verbatim by substituting these variables. Do NOT reword the shared scaffolding — reword only the variable parts. If a step needs different semantics (e.g., "re-run same skill" rather than "skip to next step"), it MUST NOT use this template; it writes the Choose block inline with its own semantics.

When NOT to use this template

• The user choice has more than two options (A/B/C/D).
• The choice is not "run-or-skip-this-skill" (e.g., "another round of the same skill", "pick tech stack", "proceed vs. rollback").
• The skipped path needs special bookkeeping beyond status: skipped (e.g., must also move artifacts, notify tracker, trigger a different skill).

For those cases, write the Choose block inline using the base format in §User Interaction Protocol.

Work Item Tracker Authentication

All tracker detection, authentication, availability gating, tracker: local fallback semantics, and leftovers handling are defined in .cursor/rules/tracker.mdc. Follow that rule — do not restate its logic here.

Autodev-specific additions on top of the rule:

Steps That Require Work Item Tracker

Before entering a step from this table for the first time in a session, verify tracker availability per .cursor/rules/tracker.mdc. If the user has already chosen tracker: local, skip the gate and proceed.

Flow	Step	Sub-Step	Tracker Action
greenfield	Plan	Step 6 — Epics	Create epics for each component
greenfield	Decompose	Step 1 + Step 2 + Step 3 — All tasks	Create ticket per task, link to epic
existing-code	Decompose Tests	Step 1t + Step 3 — All test tasks	Create ticket per task, link to epic
existing-code	New Task	Step 7 — Ticket	Create ticket per task, link to epic

State File Marker

Record the resolved choice in the state file once per session: tracker: jira or tracker: local. Subsequent steps read this marker instead of re-running the gate.

Error Handling

All error situations that require user input MUST use the Choose A / B / C / D format.

Situation	Action
State detection is ambiguous (artifacts suggest two different steps)	Present findings and use Choose format with the candidate steps as options
Sub-skill fails or hits an unrecoverable blocker	Use Choose format: A) retry, B) skip with warning, C) abort and fix manually
User wants to skip a step	Use Choose format: A) skip (with dependency warning), B) execute the step
User wants to go back to a previous step	Use Choose format: A) re-run (with overwrite warning), B) stay on current step
User asks "where am I?" without wanting to continue	Show Status Summary only, do not start execution

Failure Handling

One retry ladder covers all failure modes: explicit failure returned by a sub-skill, stuck loops detected while monitoring, and persistent failures across conversations. The single counter is retry_count in the state file; the single escalation is the Choose block below.

Failure signals

Treat the sub-skill as failed when ANY of the following is observed:

• The sub-skill explicitly returns a failed result (including blocked subagents, auto-fix loop exhaustion, prerequisite violations).
• Stuck signals: the same artifact is rewritten 3+ times without meaningful change; the sub-skill re-asks a question that was already answered; no new artifact has been saved despite active execution.

Retry ladder

Failure observed
  │
  ├─ retry_count < 3 ?
  │    YES → increment retry_count in state file
  │         → re-read the sub-skill's SKILL.md and _docs/_autodev_state.md
  │         → resume from the last recorded sub_step (restart from sub_step 1 only if corruption is suspected)
  │         → loop
  │
  │    NO (retry_count = 3) →
  │         → set status: failed and retry_count: 3 in Current Step
  │         → play notification sound (.cursor/rules/human-attention-sound.mdc)
  │         → escalate (Choose block below)
  │         → do NOT auto-retry until the user intervenes

Rules:

1. Auto-retry is immediate — do not ask before retrying.
2. Preserve sub_step across retries unless the failure indicates artifact corruption.
3. Reset retry_count: 0 on success.
4. The counter is per step, per cycle. It is not cleared by crossing a session boundary — persistence across conversations is intentional; it IS the circuit breaker.

Escalation

══════════════════════════════════════
 SKILL FAILED: [Skill Name] — 3 consecutive failures
══════════════════════════════════════
 Step: [N] — [Name]
 SubStep: [M] — [sub-step name]
 Last failure reason: [reason]
══════════════════════════════════════
 A) Retry with fresh context (new conversation)
 B) Skip this step with warning
 C) Abort — investigate and fix manually
══════════════════════════════════════
 Recommendation: A — fresh context often resolves
 persistent failures
══════════════════════════════════════

Re-entry after escalation

On the next invocation, if the state file shows status: failed AND retry_count: 3, do NOT auto-retry. Present the escalation block above first:

• User picks A → reset retry_count: 0, set status: in_progress, re-execute.
• User picks B → mark step skipped, proceed to the next step.
• User picks C → stop; return control to the user.

Incident retrospective

Immediately after the user has made their A/B/C choice, invoke .cursor/skills/retrospective/SKILL.md in incident mode:

mode: incident
failing_skill: <skill name>
failure_summary: <last failure reason string>

This produces _docs/06_metrics/incident_<YYYY-MM-DD>_<skill>.md and appends 1–3 lessons to _docs/LESSONS.md under process or tooling. The retro runs even if the user picked Abort — the goal is to capture the pattern while it is fresh. If the retrospective skill itself fails, log the failure to _docs/_process_leftovers/ but do NOT block the user's recovery choice from completing.

Context Management Protocol

Principle

Disk is memory. Never rely on in-context accumulation — read from _docs/ artifacts, not from conversation history.

Minimal Re-Read Set Per Skill

When re-entering a skill (new conversation or context refresh):

• Always read: _docs/_autodev_state.md
• Always read: the active skill's SKILL.md
• Conditionally read: only the _docs/ artifacts the current sub-step requires (listed in each skill's Context Resolution section)
• Never bulk-read: do not load all _docs/ files at once

Mid-Skill Interruption

If context is filling up during a long skill (e.g., document, implement):

1. Save current sub-step progress to the skill's artifact directory
2. Update _docs/_autodev_state.md with exact sub-step position
3. Suggest a new conversation: "Context is getting long — recommend continuing in a fresh conversation for better results"
4. On re-entry, the skill's resumability protocol picks up from the saved sub-step

Large Artifact Handling

When a skill needs to read large files (e.g., full solution.md, architecture.md):

• Read only the sections relevant to the current sub-step
• Use search tools (Grep, SemanticSearch) to find specific sections rather than reading entire files
• Summarize key decisions from prior steps in the state file so they don't need to be re-read

Context Budget Heuristic

Agents cannot programmatically query context window usage. Use these heuristics to avoid degradation:

Zone	Indicators	Action
Safe	State file + SKILL.md + 2–3 focused artifacts loaded	Continue normally
Caution	5+ artifacts loaded, or 3+ large files (architecture, solution, discovery), or conversation has 20+ tool calls	Complete current sub-step, then suggest session break
Danger	Repeated truncation in tool output, tool calls failing unexpectedly, responses becoming shallow or repetitive	Save immediately, update state file, force session boundary

Skill-specific guidelines:

Skill	Recommended session breaks
document	After every ~5 modules in Step 1; between Step 4 (Verification) and Step 5 (Solution Extraction)
implement	Each batch is a natural checkpoint; if more than 2 batches completed in one session, suggest break
plan	Between Step 5 (Test Specifications) and Step 6 (Epics) for projects with many components
research	Between Mode A rounds; between Mode A and Mode B

How to detect caution/danger zone without API:

1. Count tool calls made so far — if approaching 20+, context is likely filling up
2. If reading a file returns truncated content, context is under pressure
3. If the agent starts producing shorter or less detailed responses than earlier in the conversation, context quality is degrading
4. When in doubt, save and suggest a new conversation — re-entry is cheap thanks to the state file

Rollback Protocol

Implementation Steps (git-based)

Handled by /implement skill — each batch commit is a rollback checkpoint via git revert.

Planning/Documentation Steps (artifact-based)

For steps that produce _docs/ artifacts (problem, research, plan, decompose, document):

1. Before overwriting: if re-running a step that already has artifacts, the sub-skill's prerequisite check asks the user (resume/overwrite/skip)
2. Rollback to previous step: use Choose format:

══════════════════════════════════════
 ROLLBACK: Re-run [step name]?
══════════════════════════════════════
 A) Re-run the step (overwrites current artifacts)
 B) Stay on current step
══════════════════════════════════════
 Warning: This will overwrite files in _docs/[folder]/
══════════════════════════════════════

3. Git safety net: artifacts are committed with each autodev step completion. To roll back: git log --oneline _docs/ to find the commit, then git checkout <commit> -- _docs/<folder>/
4. State file rollback: when rolling back artifacts, also update _docs/_autodev_state.md to reflect the rolled-back step (set it to in_progress, clear completed date)

Debug Protocol

When the implement skill's auto-fix loop fails (code review FAIL after 2 auto-fix attempts) or an implementer subagent reports a blocker, the user is asked to intervene. This protocol guides the debugging process. (Retry budget and escalation are covered by Failure Handling above; this section is about how to diagnose once the user has been looped in.)

Structured Debugging Workflow

When escalated to the user after implementation failure:

1. Classify the failure — determine the category:

   • Missing dependency: a package, service, or module the task needs but isn't available
   • Logic error: code runs but produces wrong results (assertion failures, incorrect output)
   • Integration mismatch: interfaces between components don't align (type errors, missing methods, wrong signatures)
   • Environment issue: Docker, database, network, or configuration problem
   • Spec ambiguity: the task spec is unclear or contradictory

2. Reproduce — isolate the failing behavior:

   • Run the specific failing test(s) in isolation
   • Check whether the failure is deterministic or intermittent
   • Capture the exact error message, stack trace, and relevant file:line

3. Narrow scope — focus on the minimal reproduction:

   • For logic errors: trace the data flow from input to the point of failure
   • For integration mismatches: compare the caller's expectations against the callee's actual interface
   • For environment issues: verify Docker services are running, DB is accessible, env vars are set

4. Fix and verify — apply the fix and confirm:

   • Make the minimal change that fixes the root cause
   • Re-run the failing test(s) to confirm the fix
   • Run the full test suite to check for regressions
   • If the fix changes a shared interface, check all consumers

5. Report — update the batch report with:

   • Root cause category
   • Fix applied (file:line, description)
   • Tests that now pass

Common Recovery Patterns

Failure Pattern	Typical Root Cause	Recovery Action
ImportError / ModuleNotFoundError	Missing dependency or wrong path	Install dependency or fix import path
TypeError on method call	Interface mismatch between tasks	Align caller with callee's actual signature
AssertionError in test	Logic bug or wrong expected value	Fix logic or update test expectations
ConnectionRefused	Service not running	Start Docker services, check docker-compose
Timeout	Blocking I/O or infinite loop	Add timeout, fix blocking call
FileNotFoundError	Hardcoded path or missing fixture	Make path configurable, add fixture

Escalation

If debugging does not resolve the issue after 2 focused attempts:

══════════════════════════════════════
 DEBUG ESCALATION: [failure description]
══════════════════════════════════════
 Root cause category: [category]
 Attempted fixes: [list]
 Current state: [what works, what doesn't]
══════════════════════════════════════
 A) Continue debugging with more context
 B) Revert this batch and skip the task (move to backlog)
 C) Simplify the task scope and retry
══════════════════════════════════════

Status Summary

On every invocation, before executing any skill, present a status summary built from the state file (with folder scan fallback). For re-entry (state file exists), cross-check the current step against _docs/ folder structure and present any status: failed state to the user before continuing.

Banner Template (authoritative)

The banner shell is defined here once. Each flow file contributes only its step-list fragment and any flow-specific header/footer extras. Do not inline a full banner in flow files.

═══════════════════════════════════════════════════
 AUTODEV STATUS (<flow-name>)<header-suffix>
═══════════════════════════════════════════════════
<step-list from the active flow file>
═══════════════════════════════════════════════════
 Current: Step <N> — <Name><current-suffix>
 SubStep: <M> — <sub-skill internal step name>
 Retry:   <N/3>                              ← omit row if retry_count is 0
 Action:  <what will happen next>
<footer-extras from the active flow file>
═══════════════════════════════════════════════════

Slot rules

• <flow-name> — greenfield, existing-code, or meta-repo.
• <header-suffix> — optional, flow-specific. The existing-code flow appends — Cycle <N> when state.cycle > 1; other flows leave it empty.
• <step-list> — a fixed-width table supplied by the active flow file (see that file's "Status Summary — Step List" section). Row format is standardized:

   Step <N>   <Step Name>   [<state token>]
  

  where <state token> comes from the state-token set defined per row in the flow's step-list table.
• <current-suffix> — optional, flow-specific. The existing-code flow appends (cycle <N>) when state.cycle > 1; other flows leave it empty.
• Retry: row — omit entirely when retry_count is 0. Include it with <N>/3 otherwise.
• <footer-extras> — optional, flow-specific. The meta-repo flow adds a Config: line with _docs/_repo-config.yaml state; other flows leave it empty.

State token set (shared)

The common tokens all flows may emit are: DONE, IN PROGRESS, NOT STARTED, SKIPPED, FAILED (retry N/3). Specific step rows may extend this with parenthetical detail (e.g., DONE (N drafts), DONE (N tasks), IN PROGRESS (batch M of ~N), DONE (N passed, M failed)). The flow's step-list table declares which extensions each step supports.