Pipeline state

pipeline-state — the squad-wide handoff bus

Durable, file-backed state for the ENTIRE pipeline. One JSON file persists the run's variables + per-phase outputs + .ai_log/ evidence-path references across phases, so a downstream agent (or el-capitan) recovers exactly what it needs by LOOKUP instead of receiving a large inlined dump — and nothing is lost to context compression on long runs.

The standing pattern (rule R17): evidence -> .ai_log/; references + key fields -> pipeline-state; handoffs = state lookups, not inlined dumps.

• el-capitan INITIALIZES the state at session start (alongside creating the session folder), passes its path (state_file) to every dispatched agent, and after each phase READS that phase's output fields + .ai_log/ evidence paths FROM the state to validate the gate and to build the NEXT phase's prompt — passing state keys + paths, never re-dumping large content.
• every agent, on START, READS the state to pull the inputs this phase needs (key fields + .ai_log/ evidence paths) — it does NOT expect large inlined content; on COMPLETION, WRITES its outputs — key result fields + PATHS to evidence offloaded in .ai_log/ (never the blobs).
• It COEXISTS with the docs/-native reports (R7/R8 — the human-readable record) and the .ai_log/ path-handoff (R15). The state REFERENCES the docs/ + .ai_log/ artifacts; it does not replace them.

State file (PER-SESSION, .ai_log/, git-ignored)

.ai_log/session-<session-id>-<name>.json

ONE file per run, living under .ai_log/ — e.g. .ai_log/session-2026-06-30-member-blog-gating.json.

Naming convention. <session-id> (<xxxxx>) is the run's disambiguating id — in this squad the session date <YYYY-MM-DD> (distinguishes repeat/concurrent runs); <name> is the epic-slug. Together <session-id>-<name> is the session LABEL <YYYY-MM-DD>-<epic-slug> (the same slug as session_dir). The script DERIVES the path from that label (session-<label>.json).

el-capitan derives + creates it at init and passes its exact PATH as state_file to every dispatched agent. Agents read/write the path el-capitan gave them (via state_file) — they do NOT reconstruct or hardcode a constant path (the path now varies per session). This one convention is used consistently by the skill, the el-capitan-ror command, the workflow YAML, and CLAUDE.md.

Git behavior. Because it lives under .ai_log/, it is git-IGNORED (never committed) by the /.ai_log/* + !/.ai_log/.gitkeep rule — no .gitignore change needed. It is the ONE structured handoff JSON in .ai_log/; ordinary evidence there stays throwaway (R15). It is ephemeral, per-run, and regenerated fresh each session — NOT a durable record. Durable records still live in docs/ (R7/R8), which the state only references (R15/R17 stay coherent: the state is a machine handoff bus, not durable content parked in .ai_log/).

Script

.claude/skills/pipeline-state/scripts/pipeline-state.sh

Commands

# Start a new run — writes .ai_log/session-<session_id>.json AND cleans stale .ai_log/ artifacts (R15)
pipeline-state.sh init   <session_id>

# Store a string/scalar field
pipeline-state.sh set    <session_id> <field> <value>

# Append a value to a list field (creates the list if absent) — e.g. an .ai_log/ evidence path
pipeline-state.sh append <session_id> <field> <value>

# Read a single field (or the full JSON if field is omitted) — the recovery / handoff lookup
pipeline-state.sh get    <session_id> [field]

# Delete the state file
pipeline-state.sh clear  <session_id>

<session_id> is the run label on init (convention: <YYYY-MM-DD>-<epic-slug>, the same slug as session_dir); the state file PATH is DERIVED from it (.ai_log/session-<session_id>.json), so every later command passes the SAME <session_id> and resolves the SAME per-session file. el-capitan captures the path init prints and passes it as state_file; agents use that exact path.

State schema + read/write protocol

See references/state-schema.md — the per-phase entry shape (agent, status, key output fields, .ai_log/ evidence paths, work-item links) + the read-on-START / write-on-COMPLETION protocol and the field conventions (scalar via set, list via append).

init cleanup

On init the script clears stale .ai_log/ artifacts from prior runs — everything under .ai_log/ except the tracked .ai_log/.gitkeep — so a fresh run starts clean. This is consistent with R15 (.ai_log/ is temporary-only; contents are git-ignored) and with the folder-tracked/contents-ignored convention (/.ai_log/* + !/.ai_log/.gitkeep).

Usage

# --- el-capitan, at SESSION START (alongside creating docs/sessions/<session>/) ---
SESSION="2026-06-30-member-blog-gating"          # = the session_dir slug
bash .claude/skills/pipeline-state/scripts/pipeline-state.sh init "$SESSION"
bash .claude/skills/pipeline-state/scripts/pipeline-state.sh set "$SESSION" session_dir "docs/sessions/$SESSION/"
bash .claude/skills/pipeline-state/scripts/pipeline-state.sh set "$SESSION" user_request "$USER_REQUEST"
# init derived .ai_log/session-$SESSION.json — el-capitan passes state_file=.ai_log/session-$SESSION.json to EVERY dispatched agent

# --- any phase, on COMPLETION: write outputs = key fields + .ai_log/ evidence PATHS (never blobs) ---
bash .claude/skills/pipeline-state/scripts/pipeline-state.sh set    "$SESSION" prd_document "docs/prd/member-blog-gating.md"
bash .claude/skills/pipeline-state/scripts/pipeline-state.sh append "$SESSION" phase-0-evidence ".ai_log/phase-0-pm-ror-prd.md"

# --- next phase, on START: read the inputs it needs (fields + .ai_log/ paths), not a big dump ---
PRD=$(bash .claude/skills/pipeline-state/scripts/pipeline-state.sh get "$SESSION" prd_document)

# --- el-capitan, at the GATE: read the phase's outputs + evidence paths to validate + build next prompt ---
bash .claude/skills/pipeline-state/scripts/pipeline-state.sh get "$SESSION"        # full state (recovery)

e2e-test-healer note: the healer (its own out-of-pipeline flow) uses the SAME script and file — init at the start of a heal run, set/append for its per-phase variables (test id, spec file, RCA report path, burn-evidence paths, PR number), and get to restore state after context compression. It is one consumer of the now squad-wide bus; the schema in references/state-schema.md covers both the pipeline phases and the healer's fields.

---

VERSION 2.0.0