Not in the top 1,000 by GitHub stars
context4ai/agent-graph
Graph Engineering for Agent Skills: a specification and toolchain for dynamically discovering context and building observable, testable, and recoverable agent workflows.
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
examples/choice-routing/skills/chooser/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/dynamic-context/skills/inspector/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/dynamic-resource/skills/diagnose/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/fact-driven-batch/skills/operator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/getting-started/skills/operator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/independent-verification/skills/produce/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/provider-registry/consumer-a/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/provider-registry/consumer-b/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/shared-provider/skills/release/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/shared-provider/skills/release-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
examples/simple-skill/skills/draft/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
templates/basic/skills/operator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ee93426527b1 → ee93426527b1
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ee93426527b1
No earlier version was observed for comparison.
Verified commit ee93426527b1ce4dbfb750b03ee9b2965198e5ce
git clone --filter=blob:none --no-checkout https://github.com/context4ai/agent-graph.git .hotskills/context4ai-agent-graph && git -C .hotskills/context4ai-agent-graph checkout --detach ee93426527b1ce4dbfb750b03ee9b2965198e5ce
# Copy each selected skill directory from .hotskills/context4ai-agent-graph into .agents/skills/<skill-name>/Heat
2.6- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
30-day change not yet available: not enough history.
Individual skills in this repository
- choose-current-option
examples/choice-routing/skills/chooser - dynamic-context-inspector
examples/dynamic-context/skills/inspector - diagnose-from-context
examples/dynamic-resource/skills/diagnose - fact-driven-batch
examples/fact-driven-batch/skills/operator - getting-started-operator
examples/getting-started/skills/operator - produce-artifact
examples/independent-verification/skills/produce - consumer-a
examples/provider-registry/consumer-a - consumer-b
examples/provider-registry/consumer-b - release-package
examples/shared-provider/skills/release - review-package-release
examples/shared-provider/skills/release-review - draft-artifact
examples/simple-skill/skills/draft - {{providerId}}-operator
templates/basic/skills/operator
How it works
choose-current-option
When your agent uses it
Execute only the option identified by the selected Agent Graph route.
What it loads into context
Frontmatter
48 tokensAlways available to the agent
SKILL.md
75 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset154 bytesSKILL.mdcore356 bytesSKILL.mdcore356 bytes
Read SKILL.md
# Choose current option
Use the selected Route node as the option identity. Do not execute alternatives that were not selected.dynamic-context-inspector
When your agent uses it
Inspect the explicitly materialized workspace status for the current route.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
69 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset153 bytesSKILL.mdcore331 bytes
Read SKILL.md
# Inspector
Materialize and read the required context view before reporting the current state.diagnose-from-context
When your agent uses it
Diagnose a problem from the explicitly materialized context view.
What it loads into context
Frontmatter
47 tokensAlways available to the agent
SKILL.md
75 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset158 bytesSKILL.mdcore359 bytes
Read SKILL.md
# Diagnose from context
Materialize and read the required context view when it is marked `read-required` before drawing conclusions.fact-driven-batch
When your agent uses it
Process a host-owned batch through a static Agent Graph.
What it loads into context
Frontmatter
48 tokensAlways available to the agent
SKILL.md
98 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset185 bytesSKILL.mdcore453 bytes
Read SKILL.md
# Fact-driven batch
Evaluate the bound graph, resolve its primary route, read required resources marked `read-required`, and
delegate host handlers to the integration host. The host owns target selection and
refreshes Facts after every action.getting-started-operator
When your agent uses it
Prepare the current requested result using the route returned by Agent Graph.
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
80 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset194 bytesSKILL.mdcore372 bytes
Read SKILL.md
# Operator
Read the required route resources marked `read-required`, prepare the result, record the route outcome, and evaluate again.produce-artifact
When your agent uses it
Produce the requested artifact and return its observable digest.
What it loads into context
Frontmatter
18 tokensAlways available to the agent
SKILL.md
42 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
SKILL.mdcore223 bytes
Read SKILL.md
Create the artifact requested by the host. Report its location and digest so the host can observe them as facts.consumer-a
When your agent uses it
Resolve a shared Agent Graph provider through the host registry.
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
67 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
SKILL.mdcore299 bytes
Read SKILL.md
# Consumer A
Ask the host to resolve `shared.provider`, then consume the selected route.consumer-b
When your agent uses it
Use the same host-resolved provider without copying graph definitions.
What it loads into context
Frontmatter
46 tokensAlways available to the agent
SKILL.md
68 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
SKILL.mdcore305 bytes
Read SKILL.md
# Consumer B
Ask the host to resolve `shared.provider`, then consume the selected route.release-package
When your agent uses it
Guide an Agent through the shared release graph.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
66 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset185 bytesSKILL.mdcore296 bytes
Read SKILL.md
# Release package
Use the `release` graph. Never infer approval from a successful inspection.review-package-release
When your agent uses it
Review a package using the shared release provider.
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
67 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset185 bytesSKILL.mdcore299 bytes
Read SKILL.md
# Review package release
Use the same `release` graph but stop at any unresolved gate.draft-artifact
When your agent uses it
Draft an artifact using only the requirements selected by the current route.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
76 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset150 bytesSKILL.mdcore355 bytes
Read SKILL.md
# Draft artifact
Read the required route resources marked `read-required`, produce the artifact, and report an explicit outcome.{{providerId}}-operator
When your agent uses it
Execute the current Agent Graph route using only its declared resources and action contract.
What it loads into context
Frontmatter
54 tokensAlways available to the agent
SKILL.md
92 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../../provider.yamlasset178 bytesSKILL.mdcore436 bytes
Read SKILL.md
# Operator
Evaluate the graph, resolve one current route, read its required resources marked `read-required`, execute only that route, record an explicit outcome, and evaluate again.Skills included
- choose-current-option
- dynamic-context-inspector
- diagnose-from-context
- fact-driven-batch
- getting-started-operator
- produce-artifact
- consumer-a
- consumer-b
- release-package
- review-package-release
- draft-artifact
- {{providerId}}-operator
Repository facts
- License
- MIT
- Created
- Jul 22, 2026
- Last pushed
- Oct 18, 2026
- Forks
- 2
- Open issues
- 0
- Project activity
- Active
Before you install
Instructions onlyAutomated signals, not a guarantee
Checked Nov 7, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 1 note
examples/choice-routing/skills/chooser/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- No dangerous shell commands
- No network access
- No hidden or obfuscated text
- No instruction overrides
- No executable scripts
- No access to secrets
- No requests to vote in Hot
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
No combined signals at this commit.
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md examples/choice-routing/skills/chooser/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Native | Manifest AGENTS.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md examples/choice-routing/skills/chooser/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md examples/choice-routing/skills/chooser/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md examples/choice-routing/skills/chooser/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md examples/choice-routing/skills/chooser/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
