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.
npx skills add Ashkaan/contextium --skill implementgit clone --depth 1 https://github.com/Ashkaan/contextiumWrote 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.
[](https://agentmods.dev/skills/ashkaan/contextium/implement)<a href="https://agentmods.dev/skills/ashkaan/contextium/implement"><img src="https://agentmods.dev/badge/skills/ashkaan/contextium/implement/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.
<a href="https://agentmods.dev/skills/ashkaan/contextium/implement"><img src="https://agentmods.dev/badge/skills/ashkaan/contextium/implement.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 9 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 8 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 19 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 9 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 22 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 10 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 23 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 20 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium MCP Rug Pull · line 131 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium Excessive Agency · line 186 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00034 | $0.10420 |
| Opus 5.5 | $0.00014 | $0.04168 |
| Sonnet 5.5 | $0.00007 | $0.02084 |
| Haiku 4.5 | $0.00003 | $0.01042 |
Grade A, and why
implement 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 560 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/implement — Plan-Driven Executor
Critical
- Fresh-context boundary matters, and nothing enforces it. A session that wrote the SPEC and invested in its choices defends them mid-stream instead of checking them, so open a new session (or
/clear) before invoking — that is a judgment you make, not one the harness makes for you. - The worktree is mandatory. Phase 0.5 sets
${WORKTREE_DIR}; every edit MUST target absolute paths under it. The shared-checkout guard (.agents/hooks/check-shared-checkout-write.sh) refuses writes into the checkout other sessions share. Do NOT edit in the main repo. A row whose code is in ANOTHER repo (a product) adds a second worktree, of that repo, and every edit to its code and every check and review of it target that one — § "A row whose code is in another repo" below. - The check ORDER is a script, not a table. Phase 4 calls
scripts/validate.ts, which runs lint → tests → quality → review → re-lint-if-dirty in one fixed sequence. You do not choose the order and there is no argument that spells "review before lint". The table that used to live here put the code review at layer 4, ahead of E2E — so a clean review landed on a session that had run none, andaudit-dedupe.tsthen no-op'd the real review when it finally came. - E2E (Phase 4) is a hard gate, and it always runs. Static checks + unit tests are never sufficient. Re-read the spec's E2E walk (tasks.md's test tasks) and execute every step against a running app before claiming complete — once after
--phase checks, and again every time a later phase printsNEED_E2E=1. An unchanged review does not buy a skipped E2E. - If it is a UI,
/qaruns fully (Phase 4.8).validate.ts --phase qa-listsays whether a web app changed; that is a lookup, not a judgment. Any target it prints gets a complete/qa— the installed impeccable, screenshots, sight-check, visual review — and--require-qarefuses the close until every target has a marker for the tree that is about to ship. - Mechanism-match (Phase 4.5) is a hard gate. Before reporting done, write the one-line verification — agreed mechanism vs. what the diff did. If they don't match, fix the diff or surface the divergence; never paper over per "goal alignment"'s SHIP clause.
- implement-audit (Phase 4.7) is a hard gate. The built diff gets its ONE machine review via
/implement-audit— the single code reviewer for the repo, the diff-side mirror of/spec-audit. Fix every ready finding in-session; proceed only on a clean pass. This writes theimplement-audit:line the report carries. - Auto-close (Phase 6) is the loop tail. On clean completion
/implementinvokes/closeitself perclose/references/auto-close-gate.md— the user does not type/close. It HALTS instead only on a deferral or a mechanism-match FAIL. - Golden Rule: if validation fails, fix it before moving on. Never accumulate broken state.
What ships with it
22 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/templates/report.md 1.5 KB
- scripts/fixtures/layer-3-checks/check-decision-records.ts 488 B runs code
- scripts/fixtures/layer-3-checks/check-secrets.ts 114 B runs code
- scripts/fixtures/layer-3-checks/check-standards-refs.ts 195 B runs code
- scripts/layer-1.test.ts 13 KB runs code
- scripts/layer-1.ts 12 KB runs code
- scripts/layer-2.test.ts 11 KB runs code
- scripts/layer-2.ts 12 KB runs code
- scripts/layer-3.test.ts 5.8 KB runs code
- scripts/layer-3.ts 5.8 KB runs code
- scripts/peer-sweep.test.ts 4.4 KB runs code
- scripts/peer-sweep.ts 5.6 KB runs code
- scripts/resolve-scope.test.ts 18 KB runs code
- scripts/resolve-scope.ts 12 KB runs code
- scripts/session-key.test.ts 6.7 KB runs code
- scripts/session-key.ts 5.2 KB runs code
- scripts/session-write-root.test.ts 22 KB runs code
- scripts/session-write-root.ts 27 KB runs code
- scripts/setup-worktree.sh 22 KB runs code
- scripts/setup-worktree.test.ts 36 KB runs code
- scripts/validate.test.ts 41 KB runs code
- scripts/validate.ts 32 KB runs code
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.
- yesterday Changed · +14 lines 2e99b14f7fae
- 12d ago Changed · +35 lines 28ccbab8d8ec
- 15d ago Changed · +2 lines cf75c308905e
- 16d ago Changed · +298 lines · -59 tokens per session aff8068a4c27
- 17d ago Changed · -36 lines · +55 tokens per session e3e553ba4197
- 1mo ago Changed d6bdf3de02c3
- 1mo ago First seen · 247 lines · 38 tokens per session scan A d6c03cca19fd
implement is a skill published in the GitHub repository Ashkaan/contextium (36 stars, last pushed yesterday), licensed MIT. It adds 34 tokens to every session and 10,420 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
dashboard-builder
Build self-contained interactive HTML dashboards with charts, filters, and tables. Generates a single browser-openable file — no server or dependencies required.
therapeutic-ifs
Unified inner work engine: Schema deconstruction (diagnosis) + IFS therapy (treatment). Absorbs: schema-deconstruction.
bionic-decision-engine
Unified mathematical arbitrator for all resource allocation decisions — money, time, energy, relationships. Absorbs 46 decision protocols + 24 strategy protocols into one dense engine.
red-team-review
Unified adversarial review, two modes. Strategic Matrix v4.3 (MTA-004) for arguments/plans/decisions: Priors → Rubric → Adversarial Lenses → SWOT/TOWS → MCDA → Blind Spot/Kill Switch → Summary. Deliverable Audit for built artefacts (assignments, reports, code outputs): isolated fresh-context hunt against the brief…
operational-osint
Operational open-source intelligence engine for counterparty vetting, corporate registry forensics, digital footprint reconnaissance, and protective OPSEC audits.
statistical-analysis
Structured pipeline for statistical analysis deliverables — SPSS, R, Python. Covers reliability, chi-square, correlation, regression, assumption checking, and client-ready reporting.