instinct-review

instinct-review is a skill for Claude Code from softspark/ai-toolkit. It costs 36 tokens per session (1,165 once invoked), scanned A, original, Apache-2.0.

A tool for reviewing project-specific notes about how an AI coding agent should behave. These notes are stored as small Markdown files called instincts and can be kept, promoted, removed, or cleared.

In plain words
What is it for?
Use it to list instincts, review them interactively, promote a note so it is always applied, remove one, or clear all stored instincts.
Why use it?
It helps prevent useful working patterns from becoming outdated or cluttering future sessions. Reviewing them keeps the agent's project guidance relevant.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: reads .claude/ paths.

Part of the ai-toolkit plugin — 115 skills, 44 agents, 14 hooks shipped together

Good fit Use it to list instincts, review them interactively, promote a note so it is always applied, remove one, or clear all stored instincts.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/softspark/ai-toolkit/instinct-review
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add softspark/ai-toolkit --skill instinct-review
Clone the repo
git clone --depth 1 https://github.com/softspark/ai-toolkit

Made for: Claude Code.

Or install ai-toolkit, the plugin that ships this one along with the rest of its 115 skills, 44 agents, 14 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for instinct-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/softspark/ai-toolkit/instinct-review/github.svg)](https://agentmods.dev/skills/softspark/ai-toolkit/instinct-review)
Your own site
<a href="https://agentmods.dev/skills/softspark/ai-toolkit/instinct-review"><img src="https://agentmods.dev/badge/skills/softspark/ai-toolkit/instinct-review/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for instinct-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/softspark/ai-toolkit/instinct-review"><img src="https://agentmods.dev/badge/skills/softspark/ai-toolkit/instinct-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,165 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00036 $0.01165
Opus 5 $0.00018 $0.00583
Sonnet 5 $0.00007 $0.00233
Haiku 4.5 $0.00004 $0.00117

Measured 8d ago against content hash affc8d3276f7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

instinct-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

app/skills/instinct-review/SKILL.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/instinct-review - Manage Learned Instincts

$ARGUMENTS

What This Does

Manages the instinct system: project-local behavioral notes that Claude loads at the start of a session.

Instincts are markdown files in .claude/instincts/, loaded at session start by session-start.sh whenever any exist (set AI_TOOLKIT_HOOK_QUIET=1 to suppress).

Commands

List all instincts

/instinct-review --list

Shows all instincts with confidence scores and source sessions.

Review and curate

/instinct-review

Interactive review: shows each instinct and asks promote/remove/keep.

Promote instinct (always apply)

/instinct-review --promote <filename>

Remove instinct

/instinct-review --remove <filename>

Clear all instincts

/instinct-review --clear

How Instincts Work

  1. Authoring: Instinct files are written by hand — one markdown file per pattern in .claude/instincts/. There is no automatic session-end extractor today (see When NOT to Use).
  2. Storage: .claude/instincts/<pattern-name>.md, each carrying a confidence score (see format below).
  3. Loading: At session start, session-start.sh loads every instinct file into context whenever any exist. No instincts on disk means nothing is loaded; set AI_TOOLKIT_HOOK_QUIET=1 to suppress.
  4. Curation: Use /instinct-review to list, promote, remove, or clear them.

Instinct Format

# Pattern: [pattern name]
Confidence: 0.85
Sessions: 3
Last seen: 2026-03-25

[Description of the pattern or preference]

Steps

  1. Run ls .claude/instincts/ 2>/dev/null || echo "No instincts yet"
  2. For each instinct file, read it and display confidence/summary
  3. Based on $ARGUMENTS:
    • --list: display table of all instincts
    • --promote <id>: set confidence to 1.0, add "pinned" tag
    • --remove <id>: delete the file
    • --clear: delete all files in .claude/instincts/
    • no args: interactive review of each instinct
  4. Report summary of changes made

Read the full file on GitHub · 102 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 102 lines · 36 tokens per session scan A affc8d3276f7

Subscribe to this mod's changes

instinct-review is a skill published in the GitHub repository softspark/ai-toolkit (172 stars, last pushed yesterday), licensed Apache-2.0. It adds 36 tokens to every session and 1,165 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

conflict-resolver

Résolution de conflits entre agents ou sous-agents quand les résultats sont contradictoires, avec workflow de détection, arbitrage, merge et feedback loop. Se déclenche avec "conflit agent", "agent conflict", "résultats contradictoires", "agent disagreement", "contradiction agents", "arbitrage agent", "résoudre…

khalilbenaz/claude-skills-collection · 99 tokens

email-agent-builder

Construction d'agents de gestion d'emails incluant tri, réponse automatique, extraction et classification. Se déclenche avec "email agent", "agent email", "tri automatique", "réponse automatique email", "classification email. Also triggers on "auto reply agent", "inbox triage agent".

khalilbenaz/claude-skills-collection · 65 tokens

langgraph-designer

Conception de graphes d'agents avec LangGraph pour workflows complexes et stateful. Gestion d'état partagé, edges conditionnels, persistance et human-in-the-loop. Se déclenche avec "LangGraph", "graph agent", "state machine agent", "agent graph", "conditional edges", "checkpointer", "agent workflow stateful"…

khalilbenaz/claude-skills-collection · 103 tokens

pipeline-composer

Composition de pipelines de sous-agents où la sortie d'un agent alimente l'entrée du suivant. Se déclenche avec "pipeline agent", "chaîne d'agents", "agent chain", "agent pipeline", "sequential agents", "workflow agents", "ETL agent", "agent DAG", "composer agents". Also triggers on "chain agents", "output of one…

khalilbenaz/claude-skills-collection · 84 tokens

human-in-the-loop-designer

Design de systèmes human-in-the-loop pour agents IA avec approbation, correction et escalade. Se déclenche avec "human in the loop", "HITL", "approbation humaine", "validation humaine", "escalade agent", "agent supervisé", "intervention humaine", "approval workflow". Also triggers on "approval step for agent", "human…

khalilbenaz/claude-skills-collection · 88 tokens

workflow-automation-agent

Construction d'agents d'automatisation de workflows métier incluant RPA intelligent, déclencheurs et actions chaînées, orchestration d'étapes avec gestion d'erreurs et monitoring. Se déclenche avec "workflow agent", "RPA", "automatisation métier", "agent d'automatisation", "process automation. Also triggers on…

khalilbenaz/claude-skills-collection · 91 tokens