Agent skill

Context Os CLI

by jacob-dietle in jacob-dietle/context-os

This skill should be used when users ask about their work context, what they're working on, recent activity, file relationships, or knowledge graph structure.

MITAuto-check passedKnowledge Management

Install Context Os CLI

skills CLI
$ npx skills add jacob-dietle/context-os --skill context-os-cli -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install jacob-dietle/context-os context-os-cli --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/jacob-dietle/context-os.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/context-os-cli .claude/skills/context-os-cli && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
context-os-cli
GitHub stars
111
Token cost
~4.1k tokens
SKILL.md length
915 words
Files
4 (incl. references)
Skills in repo
11
Repo updated
First seen
Licence
MIT

At a glance

This skill should be used when users ask about their work context, what they're working on, recent activity, file relationships, or knowledge graph structure.

  • Tasks that involve Knowledge graphs
  • SKILL.md covers Quick Start (90% of queries), CLI Commands, The context Command and The query heat Command, plus 8 more sections
  • Calls git

What it does

Context Os CLI is an agent skill from jacob-dietle/context-os. This skill should be used when users ask about their work context, what they're working on, recent activity, file relationships, or knowledge graph structure. Uses context-os CLI (including graph health/search/stats commands) combined with grep/glob for context restoration. Every claim MUST cite a receipt ID for user verification.

Its SKILL.md is about 4.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `references/heat-metrics-model.md`, `references/query-patterns.md` and `references/search-strategies.md`).

It sits in Knowledge Management, covering Knowledge graphs. The licence is MIT.

When your agent uses it

  • Tasks that involve Knowledge graphs

Example prompts

  • “/context-os-cli”

What it can do on your machine

Read from SKILL.md and the folder at commit 1027e3f. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Context Os CLI loads about 4.1k tokens when it runs, and up to ~7.5k if it reads all its reference files. Until then it costs about 87 tokens; SKILL.md has 915 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~87
When it runs · the whole SKILL.md, loaded when a task matches
~4.1k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~7.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from jacob-dietle/context-os at commit 1027e3f, republished under its MIT licence (© jacob-dietle). 915 words, ~4,099 tokens.

Download SKILL.mdSave it as .claude/skills/context-os-cli/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
context-os-cli
description
This skill should be used when users ask about their work context, what they're working on, recent activity, file relationships, or knowledge graph structure. Uses context-os CLI (including graph health/search/stats commands) combined with grep/glob for context restoration. Every claim MUST cite a receipt ID for user verification.

Context OS CLI Skill

Use context-os CLI + grep + glob together for context restoration.

Core insight: The CLI finds WHICH files matter. Grep/Read find WHAT's in them.


Quick Start (90% of queries)

bash
# 1. Full context for a topic (start here — one command does it all)
context-os context "pixee" --format json

# 2. If you need to drill deeper, use targeted queries
context-os query flex --files "*pixee*" --agg count,recency --format json

# 3. Read hot files to understand what they are
Read: <top_file_from_step_1>

# 4. Search content IN those files
Grep: pattern="hubspot|integration" path=<directory_from_step_1>

# 5. Check git for recent changes
git log --oneline -10 -- <hot_files>

Start with context for broad questions, drop to query subcommands for targeted drilldowns.


CLI Commands

Top-Level Commands
CommandPurpose
context <query>Start here. Composed query across flex, heat, chains, sessions, timeline, co-access
query <subcommand>Targeted queries (flex, heat, chains, etc.)
intel healthCheck intel service health
intel name-chain <id>Name a chain using AI
daemon statusShow daemon sync state + registration
daemon onceRun a single sync cycle
Query Subcommands
CommandPurpose
query flex --files "*pattern*"Find files by path pattern
query flex --time 7dFind files touched in time window
query heat --time 30dFile heat metrics (specificity, velocity, score, level)
query co-access <file>Find files accessed together
query session <id>All files in a session
query sessions --time 7dSession-grouped data
query chainsList conversation chains
query timeline --time 7dTimeline data for visualization
query verify <receipt>Verify a previous query result
query receiptsList recent query receipts

Always use --format json — table output truncates long paths.


The context Command

The highest-level query. One command gives you executive summary, heat, work clusters, insights, continuity, timeline, and suggested reads.

bash
context-os context "tastematter" --format json
context-os context "pixee" --time 14d --format json

Response includes:

  • executive_summary — status, work_tempo, hot_file_count, focus_ratio
  • current_state — key_metrics, evidence (file excerpts)
  • continuity — left_off_at, pending_items, chain_context
  • work_clusters — co-accessed file groups with PMI scores
  • suggested_reads — prioritized files (with surprise flag for unexpected)
  • timeline — weekly focus areas with top files and access counts
  • insights — abandoned file detection, anomalies
  • verification — receipt_id, files/sessions/pairs analyzed
  • quick_start — commands and directory structures from matching files

When to use context vs query:

NeedUse
"What's happening with project X?"context "X"
"What files are hot right now?"query heat
"Find files matching a pattern"query flex --files
"What did I work on this week?"query flex --time 7d
"Show me session history"query sessions
"What files go with this one?"query co-access <file>

The query heat Command

Shows file heat metrics with percentile-based classification.

bash
context-os query heat --time 30d --format json
context-os query heat --files "*tastematter*" --sort specificity --format csv

Options:

-t, --time <TIME>      Long window: 30d (default), 14d, 60d, 90d
-f, --files <FILES>    File path pattern filter (glob-style)
-l, --limit <LIMIT>    Max results (default: 50)
-s, --sort <SORT>      Sort by: heat (default), specificity, velocity, name
    --format <FORMAT>  Output: table (default), json, compact, csv

Heat levels use percentile classification:

  • Top 10% = HOT, 10-30% = WARM, 30-60% = COOL, Bottom 40% = COLD

Key fields: specificity (IDF-like), velocity, heat_score, heat_level

See references/heat-metrics-model.md for formula details and interpretation tables.


Graph Traversal: graph-exec Command

Executes JavaScript against an in-memory knowledge graph (markdown files with frontmatter + wiki-links). Uses Boa JS engine — pure Rust, no Node/bun.

bash
context-os graph-exec --graph <path-to-markdown-dir> '<javascript-code>'
The codemode Object (4 functions, all synchronous)
codemode.graph_search({ pattern, scope?, maxResults? })
  pattern: string        — regex to match
  scope: 'content' | 'frontmatter' | 'both' (default: 'both')
  maxResults: number     (default: 20)
  Returns: [{ path, matches: [{ line, content }], score, frontmatter }]

codemode.graph_traverse({ start, direction?, maxDepth?, filter? })
  start: string          — node name (filename without .md, e.g. "context-engineering")
  direction: 'outbound' | 'inbound' | 'both' (default: 'outbound')
  maxDepth: number       (default: 2)
  filter: { status?, domain?, tags? }
  Returns: { nodes: [{ path, name, depth, frontmatter }], edges: [{ from, to }] }

codemode.graph_read({ path, section?, maxLines? })
  path: string           — relative path (e.g. "technical/context-engineering.md")
  section: string        — extract named heading section only
  Returns: { path, content, frontmatter, outbound_links, inbound_links }

codemode.graph_query({ filter, sort?, limit? })
  filter: { status?, domain?, tags?, name?, validated_by? }
  sort: 'name' | 'last_updated' | 'status'
  limit: number          (default: 50)
  Returns: { nodes: [{ path, name, status, domain, tags, last_updated, link_count }], total }
Code Pattern: Use IIFEs

Functions are synchronous (not async). Wrap code in an IIFE:

bash
# CORRECT — IIFE returns a value
context-os graph-exec --graph 04_knowledge_base '(() => {
  const r = codemode.graph_query({ filter: { status: "canonical" } });
  return JSON.stringify({ total: r.total, names: r.nodes.map(n => n.name) });
})()'

# WRONG — bare return is a syntax error
context-os graph-exec --graph 04_knowledge_base 'return codemode.graph_search({})'

# WRONG — async not needed (and Boa doesn't support top-level await)
context-os graph-exec --graph 04_knowledge_base 'async () => { ... }'
Bash Escaping

Keep JS simple when passing inline. Avoid backslashes and nested quotes. If complex, use a heredoc:

bash
context-os graph-exec --graph 04_knowledge_base "$(cat <<'JSEOF'
(() => {
  const hubs = codemode.graph_query({ filter: {} });
  const sorted = hubs.nodes
    .sort((a, b) => (b.link_count.outbound + b.link_count.inbound) - (a.link_count.outbound + a.link_count.inbound))
    .slice(0, 10);
  return JSON.stringify(sorted.map(n => ({
    name: n.name,
    links: n.link_count.outbound + n.link_count.inbound
  })));
})()
JSEOF
)"
Common Graph Queries
bash
# Top hubs by link count
context-os graph-exec --graph 04_knowledge_base '(() => {
  const r = codemode.graph_query({ filter: {} });
  return JSON.stringify(r.nodes.sort((a,b) => (b.link_count.outbound+b.link_count.inbound)-(a.link_count.outbound+a.link_count.inbound)).slice(0,5).map(n => ({ name: n.name, out: n.link_count.outbound, in: n.link_count.inbound })));
})()'

# All canonical nodes
context-os graph-exec --graph 04_knowledge_base '(() => {
  const r = codemode.graph_query({ filter: { status: "canonical" } });
  return JSON.stringify({ total: r.total, nodes: r.nodes.map(n => n.name) });
})()'

# Traverse outbound from a node
context-os graph-exec --graph 04_knowledge_base '(() => {
  const r = codemode.graph_traverse({ start: "context-engineering", direction: "both", maxDepth: 1 });
  return JSON.stringify(r.nodes.map(n => ({ name: n.name, depth: n.depth })));
})()'

# Find orphan nodes (no links)
context-os graph-exec --graph 04_knowledge_base '(() => {
  const r = codemode.graph_query({ filter: {} });
  const orphans = r.nodes.filter(n => n.link_count.outbound === 0 && n.link_count.inbound === 0);
  return JSON.stringify({ count: orphans.length, nodes: orphans.map(n => n.name) });
})()'

# Search then read (composition)
context-os graph-exec --graph 04_knowledge_base '(() => {
  const results = codemode.graph_search({ pattern: "taste" });
  const top = results[0];
  const detail = codemode.graph_read({ path: top.path });
  return JSON.stringify({ path: detail.path, outbound: detail.outbound_links, inbound: detail.inbound_links });
})()'
MCP Alternative

If running inside Claude Code with the codemode-graph MCP server enabled, prefer the MCP tool over the CLI. The MCP tool description is embedded and the LLM gets the API right on the first call. Functions are async in MCP mode:

js
// MCP mode (async, tool description auto-loaded)
async () => {
  const results = await codemode.graph_search({ pattern: 'context' });
  return results[0];
}
When to Use graph-exec vs MCP vs grep/glob
NeedUse
Structural queries (hubs, clusters, orphans, traversal)graph-exec or MCP
Simple file search by nameGlob
Content searchGrep
File activity/heat over timecontext-os query heat
Co-access patternscontext-os query co-access
Full project contextcontext-os context

Common Mistakes

bash
# DON'T - CLI searches file PATHS, not content
context-os query flex --files "*notification design*"  # Won't find anything
Right: Use CLI for paths, grep for content
bash
# DO - Find files first, then search content
context-os query flex --files "*alert*" --format json  # Find alert-related files
Grep: pattern="notification|digest|slack" path=<results>  # Search content
Wrong: Only use CLI
bash
# DON'T - You'll miss semantic understanding
context-os query flex --files "*hubspot*"  # Returns files but not WHAT they do
Right: Combine tools
bash
# DO - CLI narrows, Read/Grep understands
context-os query flex --files "*hubspot*" --format json
Read: <top_result>  # Understand what the file actually does
Wrong: Skip context and go straight to query
bash
# DON'T - You'll miss insights, clusters, and continuity
context-os query flex --files "*pixee*" --format json
Show full SKILL.md (364 more words)Show less
Right: Start broad, drill down
bash
# DO - Get the full picture first
context-os context "pixee" --format json
# THEN drill into specific areas if needed
context-os query heat --files "*pixee*" --format json

Combining Tools

NeedTool
Broad project contextcontext-os context "<topic>"
Find files by path patterncontext-os query flex --files
File heat/activity metricscontext-os query heat
Find files by contentGrep
Read file contentsRead
Find files by name globGlob
Check recent changesgit log

Workflow pattern:

context → full picture (summary, heat, clusters, insights)
    ↓
query heat/flex → drill into specific files or patterns
    ↓
Read → understand what files contain
    ↓
Grep → find specific concepts in content
    ↓
git log → understand evolution

Database Location

Canonical: ~/.context-os/context_os_events.db

If queries return empty results:

bash
# Check database exists
ls ~/.context-os/context_os_events.db

# Check daemon status
context-os daemon status

# Run a single sync
context-os daemon once

# Rebuild indexes if needed
context-os build-chains
context-os index-files

Citation Requirements

Every claim from query results MUST include receipt ID:

markdown
Found 147 Pixee files [q_7f3a2b]
To verify: context-os query verify q_7f3a2b

When This Skill Helps vs Doesn't

HelpsDoesn't Help
"What files did I touch for project X?""What was I thinking about?"
"When was this file last accessed?""Why did I make this change?"
"What files are related to this one?""What's the best architecture for X?"
"What's the status of this work?""How should I fix this bug?"
"What files are hot/cold right now?""What's in my calendar?"

For semantic understanding, READ the files. The CLI tells you which ones matter.


References

For advanced patterns:

  • references/heat-metrics-model.md - Heat formula: specificity, exponential decay, percentile classification
  • references/search-strategies.md - 9 multi-step search strategies (Pilot Drilling, Triangulation, etc.)
  • references/query-patterns.md - Path substring patterns and result interpretation

CLI Full Reference

context (Composed Query — Start Here)
bash
context-os context <QUERY> [OPTIONS]

ARGUMENTS:
  <QUERY>            Search query (used as glob pattern *query*)

OPTIONS:
  -t, --time <TIME>      Time window (default: 30d)
  -l, --limit <LIMIT>    Max results per sub-query (default: 20)
      --format <FORMAT>  Output: json (default), compact, table
query flex (Targeted File Query)
bash
context-os query flex [OPTIONS]

OPTIONS:
  -f, --files <FILES>      File pattern (glob): "*pixee*", "*.py"
  -t, --time <TIME>        Time window: 7d (default), 14d, 30d
  -c, --chain <CHAIN>      Filter by chain ID
  -s, --session <SESSION>  Filter by session ID
  -a, --agg <AGG>          Aggregations: count, recency
  -l, --limit <LIMIT>      Max results (default: 20)
      --sort <SORT>        Order: count (default), recency
      --format <FORMAT>    Output: json (default), compact
query heat (File Heat Metrics)
bash
context-os query heat [OPTIONS]

OPTIONS:
  -t, --time <TIME>      Window: 30d (default), 14d, 60d, 90d
  -f, --files <FILES>    File pattern filter (glob-style)
  -l, --limit <LIMIT>    Max results (default: 50)
  -s, --sort <SORT>      Sort: heat (default), specificity, velocity, name
      --format <FORMAT>  Output: table (default), json, compact, csv
query timeline (Visualization Data)
bash
context-os query timeline [OPTIONS]

OPTIONS:
  -t, --time <TIME>      Time range: 7d (default), 14d, 30d
  -p, --files <FILES>    File pattern filter
  -c, --chain <CHAIN>    Filter by chain ID
  -l, --limit <LIMIT>    Max files (default: 30)
      --format <FORMAT>  Output: json (default), compact
query sessions (Session-Grouped)
bash
context-os query sessions [OPTIONS]

OPTIONS:
  -t, --time <TIME>      Time range: 7d (default), 14d, 30d
  -c, --chain <CHAIN>    Filter by chain ID
  -l, --limit <LIMIT>    Max sessions (default: 50)
      --format <FORMAT>  Output: json (default), compact
Other Query Commands
CommandPurpose
query search <term>Keyword search in file paths
query file <path>History for specific file
query co-access <path>Files accessed with target
query chainsList conversation chains
query verify <receipt>Verify a receipt
query receiptsList recent receipts

All support --format json.

Non-Query Commands
CommandPurpose
sync-gitSync git commits from repository
parse-sessionsParse JSONL session files
build-chainsBuild chain graph from sessions
index-filesBuild inverted file index
watchWatch directory for file changes
daemon onceRun single sync cycle
daemon startStart daemon (foreground)
daemon statusShow sync state + registration
daemon installInstall daemon to run on login
daemon uninstallRemove daemon from login
intel healthCheck intel service health
intel name-chain <id>Name a chain using AI
serveStart HTTP API server

Last Updated: 2026-04-07 Version: 6.0 (Renamed from context-query to context-os-cli. All local CLI commands now reference context-os binary. Publishing commands remain tastematter. Added graph health/search/stats from Spec 32.)

© jacob-dietle, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (references) in .claude/skills/context-os-cli of jacob-dietle/context-os.

  • SKILL.md
  • references/heat-metrics-model.md
  • references/query-patterns.md
  • references/search-strategies.md

Open the folder on GitHubat commit 1027e3f

Compare with similar skills

Context Os CLI next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Context Os CLI compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Context Os CLI this skilljacob-dietle/context-os111—~4.1kAutomated safety check: PassMIT
LLM Wiki Knowledge GraphEgonex-AI/Understand-Anything86k1 repos~1.5kAutomated safety check: PassMIT
Ontology1mancompany/OneManCompany4402 repos~1.5kAutomated safety check: PassApache-2.0
Graphagenticnotetaking/arscontexta3.5k1 repos~4.9kAutomated safety check: NotesMIT
Obsidian Canvas BoardsAgriciDaniel/claude-obsidian15k—~1.4kAutomated safety check: PassMIT
Knowledge Graphgnomeria/usbtree690—~1.5kAutomated safety check: PassMIT

Similar skills

  • LLM Wiki Knowledge Graph

    Egonex-AI/Understand-Anything

    Detects a Karpathy-pattern LLM wiki and builds an interactive knowledge graph with entities, implicit relationships and topic clusters.

    86k GitHub starsUsed in 1 repo~1.5k tokens
    Knowledge ManagementAuto-check passed
  • Ontology

    1mancompany/OneManCompany

    Typed knowledge graph for structured agent memory and composable skills.

    440 GitHub starsUsed in 2 repos~1.5k tokens
    Knowledge ManagementAuto-check passed
  • Graph

    agenticnotetaking/arscontexta

    Interactive knowledge graph analysis. An agent skill from agenticnotetaking/arscontexta.

    3.5k GitHub starsUsed in 1 repo~4.9k tokens
    Knowledge ManagementAuto-check: notes
  • Obsidian Canvas Boards

    AgriciDaniel/claude-obsidian

    Creates, inspects and updates Obsidian JSON Canvas boards in a vault, with text, file, link, group and edge nodes, using safe recoverable edits.

    15k GitHub stars~1.4k tokensUpdated 27 days ago
    Knowledge ManagementAuto-check passed
  • Knowledge Graph

    gnomeria/usbtree

    Set up and maintain a lightweight, file-based knowledge graph of the repo — entities, typed relations, decisions, gotchas — so agents load context fast instead of re-exploring the codebase every…

    690 GitHub stars~1.5k tokensUpdated 1 mo ago
    Knowledge ManagementAuto-check passed
  • Stats

    agenticnotetaking/arscontexta

    Show vault statistics and knowledge graph metrics. An agent skill from agenticnotetaking/arscontexta.

    3.5k GitHub starsUsed in 1 repo~3.1k tokens
    Knowledge ManagementAuto-check: notes

More from jacob-dietle/context-os

All 11 skills in this repo
  • Bottleneck Attack

    jacob-dietle/context-os

    A skill your agent uses when deciding what to work on next, when progress is stuck, or when the reflex is to build or automate before proving the current bottleneck.

    111 GitHub stars~2.3k tokensUpdated 1 mo ago
    Auto-check passed
  • Code Service Defrag

    jacob-dietle/context-os

    This skill should be used to periodically defragment a multi-app/multi-service codebase — both CODE (duplicate deploy targets, colliding bindings, stale forks) and CONTEXT (parallel spec…

    111 GitHub stars~3.7k tokensUpdated 1 mo ago
    Auto-check passed
  • Content Strategy And Assembly

    jacob-dietle/context-os

    This skill should be used when producing content (newsletter posts, blog posts, LinkedIn posts) from existing corpus material.

    111 GitHub stars~3.1k tokensUpdated 1 mo ago
    Auto-check passed
  • Decision Accountability

    jacob-dietle/context-os

    This skill should be used when making architectural decisions, writing specs, or reviewing decisions that contain "future work", "v2", "simpler for now", "out of scope", or complexity claims.

    111 GitHub stars~1.9k tokensUpdated 1 mo ago
    Auto-check passed
  • Coordinated Agent Teams

    jacob-dietle/context-os

    This skill should be used when decomposing a spec into a multi-agent implementation plan with dependency ordering, parallelism decisions, contract testing, and verification strategy.

    111 GitHub stars~4.2k tokensUpdated 1 mo ago
    Auto-check passed
  • Eval Loop

    jacob-dietle/context-os

    This skill should be used when a specific quality problem (UX, data, architecture, feature) needs systematic diagnosis and iterative fixing toward a defined target.

    111 GitHub stars~5.2k tokensUpdated 1 mo ago
    Auto-check passed

Questions about Context Os CLI

What does Context Os CLI do?

This skill should be used when users ask about their work context, what they're working on, recent activity, file relationships, or knowledge graph structure. Context Os CLI is an agent skill from jacob-dietle/context-os. This skill should be used when users ask about their work context, what they're working on, recent activity, file relationships, or knowledge graph structure.

When should I use Context Os CLI?

Context Os CLI fits situations like: tasks that involve Knowledge graphs.

How do I install Context Os CLI in Claude Code?

Run `npx skills add jacob-dietle/context-os --skill context-os-cli -a claude-code`. Or copy the skill folder (.claude/skills/context-os-cli in jacob-dietle/context-os) into .claude/skills/context-os-cli in your project. Claude Code loads it when a task matches its description.

How do I install Context Os CLI in Codex?

Run `npx skills add jacob-dietle/context-os --skill context-os-cli -a codex`. Or copy the skill folder (.claude/skills/context-os-cli in jacob-dietle/context-os) into .agents/skills/context-os-cli in your project. Codex loads it when a task matches its description.

Can I use Context Os CLI in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jacob-dietle/context-os --skill context-os-cli -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/context-os-cli, .gemini/skills/context-os-cli, .github/skills/context-os-cli and .opencode/skills/context-os-cli in your project.

What does Context Os CLI need to run?

Going by SKILL.md and its folder, Context Os CLI needs the command-line tools its instructions call (git).

Does Context Os CLI access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Context Os CLI safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Context Os CLI use?

Context Os CLI is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Context Os CLI use?

About 4.1k tokens (SKILL.md is roughly 16k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 3.4k tokens, read only when the agent opens those files.

What are the alternatives to Context Os CLI?

Skills that share tags, products or a category with Context Os CLI: LLM Wiki Knowledge Graph (Egonex-AI/Understand-Anything, 86k stars), Ontology (1mancompany/OneManCompany, 440 stars), Graph (agenticnotetaking/arscontexta, 3.5k stars) and Obsidian Canvas Boards (AgriciDaniel/claude-obsidian, 15k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Context Os CLI?

jacob-dietle (a GitHub user) maintains it in jacob-dietle/context-os, which has 111 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on August 13, 2026.

Source: jacob-dietle/context-os on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.