Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Df Context Store

ASecurity

Dark Factory Stage 0.5 — scaffold a repo's in-repo agent inner-loop substrate (read-first context store + single-responsibility agents + checkable hand-off contracts + commit gate) from the proven template, populate it from the real codebase, and verify it. Use when bootstrapping a repo for DF work, running a DF build, or adding the agent substrate to an existing repo. Triggers on "context store", "agent substrate", "inner loop", "bootstrap agents", "df stage 0.5".

3 stars
0 votes
0 copies
0 views
Added 9/24/2026
ai-agentsgobashnodegitapi

Works with

api

Security Analysis

A100/100

Pro scans all 20 files and shows the line behind each finding

Scanned 9/24/2026

$npx -y skills add OneDro1d/dark-factory --skill df-context-store --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Df Context Store?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Df Context Store
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/onedro1d-df-context-store/badge)](https://www.skillsdirectory.com/skills/onedro1d-df-context-store)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: df-context-store
description: Dark Factory Stage 0.5 — scaffold a repo's in-repo agent inner-loop substrate (read-first context store + single-responsibility agents + checkable hand-off contracts + commit gate) from the proven template, populate it from the real codebase, and verify it. Use when bootstrapping a repo for DF work, running a DF build, or adding the agent substrate to an existing repo. Triggers on "context store", "agent substrate", "inner loop", "bootstrap agents", "df stage 0.5".
allowed-tools: Read, Write, Edit, Bash, Grep, Glob
---

# df-context-store — plant the inner-loop substrate (DF Stage 0.5)

Goal: give the repo a **persistent, in-repo, read-first agent substrate** so analysis
is done once and reused, hand-offs can't silently degrade, and nothing ships
unverified. This is the inner-loop complement to the heavy DF build stages — the
build injects rigor at the seams; this is what *lives in the repo* and binds every
future session. It runs between Stage 0 (Frame) and Stage 1 (PO), and again on
demand to refresh.

Template (source of truth): the `substrate-template/` directory bundled alongside
this skill (`skills/df-context-store/substrate-template/`).

## Procedure

1. **Emit the payload.** Copy `substrate-template/dotclaude/` into the target repo
   as `.claude/` (agents + context store + `hooks/` + `contract-check` + `commit-sync`
   + `AGENT-CONTRACTS.md`). Don't overwrite an existing populated store — merge.
2. **Populate from source, don't invent.**
   - `.claude/context/SERVICE-MAP.md` — the structural map (services/db/queue/API +
     flows), generated from the real codebase (use `service-mapper`).
   - `.claude/context/DATA-FLOW.md` — the data-transform view: data nodes
     (schema/origin/**authority**/class), the transform graph (`pure|effect`), and
     the validation rules (`LOCAL` reject / `GLOBAL` reconcile). Populate nodes +
     authority from source; leave rules you can't yet cite as marked skeletons.
   - Leave `FINDINGS.md`/`DECISIONS.md` as clean skeletons that accumulate as the
     repo is worked. Fill per-service `<service>/CLAUDE.md` stubs only where verified
     (`file:line`).
3. **Wire the always-on carrier.** Merge `substrate-template/CLAUDE-stanza.md` into
   the repo's root `CLAUDE.md` (read-first/dispatch/verify rules) — this is what binds
   every session, not the one-time scaffold.
4. **Arm the lockstep gate + make it self-arming (the every-commit half).** The gate
   can't drift code from the context store silently:
   - `chmod +x .claude/hooks/pre-commit .claude/hooks/ensure-gate.sh`
   - **Self-arm carrier:** merge the `SessionStart` hook from `substrate-template/dotclaude/settings.json`
     into the repo's `.claude/settings.json` (create it if absent; if it exists,
     add the `SessionStart` entry — do **not** clobber existing hooks). This runs
     `ensure-gate.sh` on every session so a fresh clone activates the gate with no
     manual step. (`core.hooksPath` is local git config and never travels in a commit —
     the SessionStart hook is what carries the every-clone half.)
   - **Arm it now (non-destructive):** if the repo has **no** existing `core.hooksPath`
     and **no** husky, run `git config core.hooksPath .claude/hooks` (or just run
     `bash .claude/hooks/ensure-gate.sh`, which does exactly this safely). If one
     already exists, do **not** overwrite it — install the gate as
     `.claude/hooks/pre-commit.local` under the existing path (it chains). Say which
     path you took.
5. **Verify (acceptance).** Run `bash substrate-template/verify-substrate.sh`
   (from this skill dir) against the template, and for an emitted repo confirm: required files present, no
   leftover template placeholders in `SERVICE-MAP`/`DATA-FLOW` (they're populated), the
   `contract-check` suite is green, the gate is executable + actually blocks an
   undocumented structural change, and the CLAUDE.md stanza is merged. Green is the
   bar; report the literal check counts.

## Outputs
- `.claude/{agents,context,hooks,skills/contract-check,skills/commit-sync}` in the target repo.
- A read-first stanza merged into the repo's `CLAUDE.md`.
- A populated `SERVICE-MAP.md` + `DATA-FLOW.md`; clean `FINDINGS.md`/`DECISIONS.md` skeletons.
- An armed `pre-commit` staleness gate (`core.hooksPath` set, or chained via `.local`),
  self-arming on future clones via the `SessionStart` → `ensure-gate.sh` hook merged
  into `.claude/settings.json`.

## Rules
- **Populate, don't fabricate** — `SERVICE-MAP` entries cite real code; unknowns are
  marked, not guessed (same discipline the agents enforce).
- **Idempotent** — re-running merges/refreshes; it must not duplicate entries or
  clobber an accumulated `FINDINGS`/`DECISIONS` ledger.
- **Generic-in, specific-out** — the template is project-neutral; the emitted
  substrate is tailored to THIS repo. Don't ship template placeholders into a repo.
- Referenced by `dark-factory-build` as Stage 0.5. An organisation layer may also call this
  as the substrate step of its own cold-start generator where that toolchain is present —
  optional integration, not required.

Attribution

OneDro1dOneDro1d
View sourceSee grades on GitHubMore from OneDro1d →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →