Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add eigenlegal/counsel-os/plugin install counsel-osWrote 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/eigenlegal/counsel-os/counsel)<a href="https://agentmods.dev/skills/eigenlegal/counsel-os/counsel"><img src="https://agentmods.dev/badge/skills/eigenlegal/counsel-os/counsel/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/eigenlegal/counsel-os/counsel"><img src="https://agentmods.dev/badge/skills/eigenlegal/counsel-os/counsel.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00160 | $0.05855 |
| Opus 5 | $0.00080 | $0.02927 |
| Sonnet 5 | $0.00032 | $0.01171 |
| Haiku 4.5 | $0.00016 | $0.00585 |
Grade A, and why
counsel scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 366 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Counsel OS
How This System Works
Counsel OS is a legal practice system. The user says what they need — review a contract, check a clause, draft a response, look something up. You figure out how to do it using the primitives below. There are no phases, no pipeline, no slash commands for legal work. Just capabilities you compose based on the request.
Runtime hand-off (Claude Code only). Before any other step, run bash "${CLAUDE_PLUGIN_ROOT}/scripts/runtime_step.sh" "<the user's request, verbatim>". Exit 0 means the local counsel-os runtime handled the request: relay its output to the user and stop. Exit 3 means no runtime is running: continue with this skill exactly as below. Exit 1 means the runtime hit an error: show the message and continue with this skill.
The 5 Primitives
These are your capabilities. For each request, decide which ones to use and in what combination. Read the full primitive file (primitives/{name}.md) when you need the detailed instructions for a mode.
read — Understand a document. Extract structure, terms, parties, key provisions.
research — Find context from the knowledge system. Law areas, positions, entities, matters, library.
evaluate — Assess against standards and/or law. Per-issue, with rationale and citations.
draft — Generate any output. Counter-language, memos, emails, summaries, redlines.
remember — Persist state and propose knowledge updates. Matter files, entity files, practice.
Intent Routing
Match the user's request to primitives. This is guidance, not rigid rules — use judgment.
| User says | Primitives |
|---|---|
| "Review this [document]" | read → research → evaluate → draft --summary → remember --matter |
| "Review these [N documents / this folder]" | evaluate --batch (loops read + evaluate per doc → one consolidated report) → remember --matter |
| "Is this [clause] ok?" | research --position → evaluate |
| "What's our position on [topic]?" | research --position |
| "What did we agree with [counterparty]?" | research --entity |
| "Draft a response to [specific issue]" | research --entity --library → draft --counter-language |
| "Summarize this for [person/team]" | read (if needed) → draft --summary --for [audience] |
| "QA this / check the references, definitions, exhibits" | read --qa (deterministic mechanical checker) |
| "Create a redline" | read → research → evaluate → draft --redline → remember --matter |
| "They sent back a redline / what did they change?" | read --redline → research → evaluate → draft --counter-language → remember --matter |
| "Is this [document] compliant with [regulation]?" | read → research --law → evaluate --compliance |
| "Update the [counterparty] file" | remember --entity |
| "Close this matter" | remember --matter (stage → closed) → remember --entity → remember --knowledge |
| "What's due / any deadlines coming up / anything expiring?" | /counsel-os:docket (read-only sweep of deadlines: frontmatter across matters) |
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.
- 9d ago Changed af9aa53ba467
- 12d ago First seen · 366 lines · 160 tokens per session scan A a8947f27720f
counsel is a skill published in the GitHub repository eigenlegal/counsel-os (24 stars, last pushed yesterday), licensed MIT. It adds 160 tokens to every session and 5,855 once invoked, about $0.0008 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 2026-08-30.
Other skills, from other repositories
specification-writing
A workflow for writing complete patent specifications from patent claims and an invention disclosure. It adapts the document to a chosen jurisdiction, such as the US, Europe, or China.
regulatory-research-fallback
Fallback workflow for regulatory research when web extraction tools fail on government PDFs.
x-scorecard
OpenSSF Scorecard for assessing open source project security. Check security best practices and compliance. Dependency: This is an x-cmd module. Install x-cmd first (see x-cmd skill for installation options). see x-cmd skill for installation.
gesellschaftsrechtliche-satzungen-agb
Für Gesellschaftsrechtliche Satzungen AGB Abgrenzung: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt. Fachgebiet: AGB-Recht-Prüfer. Route: gesellschaftsrechtliche-satzungen-agb.
memstack-business-gdpr
Use this skill when the user says 'GDPR', 'data protection', 'privacy compliance', 'DPA', 'DSAR', 'data subject request', 'cookie consent', 'privacy audit', 'CCPA', or asks 'do I need GDPR for this repo'. Scans the repository to detect what personal data is collected, classifies sensitivity, determines whether GDPR…
nda-review
Use when the user uploads or pastes a non-disclosure agreement and asks for review, redline, risk assessment, or a recommendation on whether to sign. Identifies missing standard protections, one-sided or unusual provisions, and operational issues; produces a structured report with severity ratings and citations to…