capture-insight

name	capture_insight
description	Captures a quick insight, pattern, or discovery to the daily insights scratchpad without interrupting the current task. Use for: /capture_insight, 'note this', 'remember this pattern', 'log this decision', 'save this as a lesson', 'note that', 'remember that'. Appends a structured entry to .workflow_artifacts/memory/daily/insights-<date>.md.
model	haiku

Capture Insight

You log a single insight to the daily scratchpad immediately and return control to the user.

Portable intent doc: quoin/core/skills/capture_insight.md

§0 Model dispatch (FIRST STEP — execute before anything else)

This skill is declared model: haiku. If the executing agent is running on a model strictly more expensive than the declared tier, you MUST self-dispatch before doing the skill's actual work.

Detection:

• Read your current model from the system context ("powered by the model named X").
• Tier order: haiku < sonnet < opus.
• Sentinel parsing: the user's prompt is checked for the [no-redispatch] family.
  • Bare [no-redispatch] (parent-emit form AND user manual override): skip dispatch, proceed to §1 at the current tier.
  • Counter form [no-redispatch:N] where N is a positive integer ≥ 2: ABORT (see "Abort rule" below).
  • Counter form [no-redispatch:1] is reserved and treated as bare [no-redispatch] for forward-compatibility; do not emit it.
• If current_tier > declared_tier AND prompt does NOT start with any [no-redispatch] form: Dispatch reason: cost-guardrail handoff. dispatched-tier: haiku.

Pre-dispatch 1M check (IVG-90 Layer 1+2):

• Run: python3 QUOIN_HOME/scripts/dispatch_config.py --decide --tier <declared_tier> --verbose where <declared_tier> is the tier declared for this skill (e.g. "sonnet" or "haiku", as shown in the dispatched-tier line immediately above).
• If the command returns "safe-path" on line 1: Read the reason token from line 2 (config|cache|probe). Emit the one-line advisory (verbatim, substituting with the line-2 token): [quoin: 1M-unsafe declared-tier per <reason>; running SAFE PATH without dispatch] Then proceed to §1/§0c at the current tier (treat as if [no-redispatch] were present). Do NOT call the Agent dispatch. Do NOT call AskUserQuestion.
• If the command returns "dispatch" on line 1, OR if the script is missing / errors: Continue to the Agent dispatch call below (today's path — fail-OPEN).

  Spawn an Agent subagent with the following arguments:
    model: "haiku"
    description: "capture_insight dispatched at haiku tier"
    prompt: "[no-redispatch]\n<original user input verbatim>"
  Wait for the subagent.

  Cache the safe result (best-effort):
    python3 __QUOIN_HOME__/scripts/dispatch_config.py --write-cache --tier <declared_tier> --result safe
  (Fail-OPEN: if the script errors or is missing, silently skip and continue.)

  Return its output as your final response. STOP.
  (Return the subagent's output as your final response.)

Abort rule (recursion guard):

• If the prompt starts with [no-redispatch:N] AND N ≥ 2: ABORT before any tool calls.
• Print the one-line error: Quoin self-dispatch hard-cap reached at N=<N> in capture_insight. This indicates a recursion bug; aborting before any tool calls. Re-invoke with [no-redispatch] (bare) to override.
• Then stop. Do NOT proceed to §1.

Manual kill switch:

• The user can prefix any user-typed slash invocation with bare [no-redispatch] to skip dispatch entirely (e.g., [no-redispatch] /capture_insight).
• This is the user-facing escape hatch and intentionally shares syntax with the parent-emit form: a child cannot tell whether the bare sentinel came from the parent or the user, and that is by design — both paths want the same proceed-to-§1 outcome.
• Use this only when intentionally overriding the cost guardrail (e.g., for one-off debugging on a different tier).

Fail-graceful path with error-class triage (per architecture I-01):

• If the Agent tool returns an error during dispatch, classify the error message text BEFORE proceeding:

• Error classification:

  • Worktree-class: the error text contains the substring Cannot create agent worktree, OR (the substring worktree AND the substring not in a git repository). This is recoverable — the harness tried to create a git worktree for isolation and the project root is not a git repo. Continue to Worktree-class branch.
  • Other-class: any other tool error, exception, or harness rejection — skip to Other-class path below (existing fail-OPEN behavior).

• 1M-credit-class: if the error text contains the substring Usage credits required for 1M context: This is the 1M-context credit mismatch (IVG-89). The parent session carries the context-1m-2025-08-07 beta header which propagates to all subagent calls; the declared-tier model lacks 1M credits. Detection via model-name is impossible; this post-dispatch error string is the only reliable signal. Emit (verbatim): [quoin: 1M-context credit mismatch on <tier> subagent dispatch; proceeding in-session at parent tier — run /model to switch this session to standard context for a permanent fix]

  Cache the unsafe result (best-effort):
    python3 __QUOIN_HOME__/scripts/dispatch_config.py --write-cache --tier <declared_tier> --result unsafe
  (Fail-OPEN: if the script errors or is missing, silently skip and continue.)

  Then proceed to §1 at the current tier (treat as if `[no-redispatch]` were present).
  Do NOT retry the Agent dispatch. Do NOT call AskUserQuestion.

• Worktree-class branch: Worktree creation is hook-driven and cannot be skipped by omitting a parameter. Use the AskUserQuestion tool to present the user with one option: (c) proceed-current-tier — Skip dispatch, proceed at the current (more expensive) tier. This is the only available recovery path. Question header: Subagent dispatch failed (worktree creation). Proceeding at current tier. Note for the user: "Worktree dispatch failed and no retry mechanism is available — worktree creation is unconditional in this harness. Proceeding at current tier."

• Other-class path (also: worktree-class after user acknowledges c): Do NOT abort the user's invocation. Emit the bare warning (verbatim): [quoin-stage-1: subagent dispatch unavailable; proceeding at current tier] If this path was reached via a worktree-class error, ALSO emit the classification line (second, separate): [quoin-stage-1: error-class=worktree; user-choice=c; proceeding at current tier] Then proceed to §1 at the current tier (fail-OPEN per I-01).

Otherwise (already at or below declared tier, OR prompt has [no-redispatch] sentinel, OR dispatch unavailable): proceed to §1 (skill body).

Session bootstrap

Cost tracking note: /capture_insight is a lightweight note-taking skill. Append to the cost ledger only if a task context is clearly active. If in doubt, skip cost recording.

If a task context is active: append your session to .workflow_artifacts/<task-name>/cost-ledger.md (see cost tracking rules in CLAUDE.md) — phase: capture-insight.

Process

Step 1: Determine the insight

The insight is whatever the user just described, or what you just noticed during task work. If the user invoked this explicitly (e.g., /capture_insight "this API returns 200 on auth failure"), use that. If invoked from context (e.g., the user said "note that"), extract the key point.

Step 2: Classify it

Choose the most fitting type:

• pattern — a recurring structure in this codebase worth remembering
• gotcha — something that caused confusion or wasted time; a trap to avoid
• decision-rationale — a choice made whose "why" will be forgotten by tomorrow
• surprise — something that behaved unexpectedly
• workflow-friction — something about the workflow process itself that felt wrong or slow (Tier 3 candidate)

And set Promote?:

• yes — clearly reusable across future sessions or tasks
• maybe — worth reviewing at end of day
• no — local/ephemeral, not worth carrying forward

For workflow-friction type, always set Promote?: yes — these are surfaced as Tier 3 suggestions at /end_of_day.

Step 3: Write to scratchpad

V-05 reminder: T-NN/D-NN/R-NN/F-NN/Q-NN/S-NN are FILE-LOCAL.

When referring to a sibling artifact's task or risk, use plain English (e.g., "the parent plan's T-04"), NOT a bare T-NN token. See format-kit.md §1 / glossary.md.

Write daily/insights-{date}.md appends in v3 format per the §5.4 Class A writer mechanism. Reference files (apply HERE at the body-generation write-site, per format-kit.md §1 / lesson 2026-04-23): __QUOIN_HOME__/memory/format-kit.md (primitives + section set), __QUOIN_HOME__/memory/glossary.md (status glyphs + abbreviation whitelist), __QUOIN_HOME__/memory/terse-rubric.md (prose discipline inside the Insight body). Entry structure: each insight is a ## {HH:MM} — {task-context or "ad-hoc"} heading followed by a YAML-style key/value block (Type / Insight / Applies to / Promote?) per the template below. The terse-rubric applies inside the multi-sentence Insight value only (caveman prose for "why" content). After appending, run python3 __QUOIN_HOME__/scripts/validate_artifact.py {insights-path} (auto-detection — file does not match a named-type prefix, falls through to default). On V-failure (rare for append-only structured entries): log a format-kit-skipped warning and continue (capture_insight is Haiku, must not block). No retry — append-only file already has the new entry.

Append to .workflow_artifacts/memory/daily/insights-<YYYY-MM-DD>.md (today's date). Create the file with a header if it doesn't exist yet:

# Daily Insights — <YYYY-MM-DD>

Scratchpad for patterns, gotchas, and decisions captured during task work.
Reviewed and promoted by /end_of_day.

---

Then append the entry:

## <HH:MM> — <task-context or "ad-hoc">
**Type:** <pattern | gotcha | decision-rationale | surprise | workflow-friction>
**Insight:** <the observation, 1-3 sentences>
**Applies to:** <skill names, technology, or "general" — or "workflow" for Tier 3>
**Promote?:** <yes | maybe | no>

Step 4: Confirm

Reply with one short sentence confirming what was captured. Then stop — don't elaborate.

Example: "Noted: API silently returns 200 on auth failure (gotcha, promote: yes)."

Important behaviors

• Fast. This skill should complete in seconds. No analysis, no follow-up questions.
• Don't ask for clarification. Make a reasonable judgment about type and Promote? and proceed. The user can always delete or edit the entry.
• Don't break the user's flow. One-line confirmation only.