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 agentmods add skills/kanetran29/uigraph/kitnpx skills add kanetran29/uigraph --skill kitgit clone --depth 1 https://github.com/kanetran29/uigraphWhat 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 | $0.00044 | $0.00619 |
| Opus 5 | $0.00022 | $0.00309 |
| Sonnet 5 | $0.00009 | $0.00124 |
| Haiku 4.5 | $0.00004 | $0.00062 |
Grade A, and why
uigraph 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 2d 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 — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
uigraph agent kit
uigraph is a self-mapping behavioral graph of a UI: screens (nodes) and the transitions between them (edges), extracted statically from the source, enriched by quarantined LLM proposals, and confirmed by runtime observation. The core is model-free — you (the LLM) are a consumer, not part of the trusted base.
The one law you must not break
Read rules/00-golden-invariant.md. In one line: no edge enters the proven graph without a deterministic static or runtime witness. You never write a proven edge. The observation enters the graph, not your guess.
Vocabulary (6 lines)
- modality —
must(always fires) ·may(conditional/guarded) ·unknown(dynamic target, undecidable statically). See rules/01-modality.md. - source / provenance —
static(extractor) ·manual(human overlay) ·runtime(observed) ·proposal(quarantined lead). See rules/02-provenance.md. - a proposal is a LEAD, not a fact — see rules/03-proposal-is-a-lead.md.
The tools at a glance
Full reference: guides/00-tools.md. Grouped:
- Read — get_graph, get_proposals, get_grounding, get_proposal_graph, describe_screen, get_coverage, next_to_verify, get_loop_status
- Plan — plan_path, gen_spec, list_scenarios, set_scenario
- Mutate — update_graph, propose (Tier-2: submit greedy quarantined hypotheses), report_observation (Tier-3: confirmed requires evidence + reportedBy), reconcile_proposals, withdraw_proposal, mark_unverifiable
- Compare — diff
The job you are usually here to do
Drive proposals + uncertain edges to resolution: the reconciliation loop,
loop/reconciliation-loop.md. Read state with
get_loop_status, verify the worklist, archive what's confirmed, withdraw what's
hallucinated, until loopDone.
Reading the app's runtime state
guides/02-reading-state.md (coverage + worklist) and guides/01-verify-flow.md (Tier-3 + authenticated runs).
What ships with it
10 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.
- guides/00-tools.md 6.3 KB
- guides/01-verify-flow.md 2.6 KB
- guides/02-reading-state.md 1.7 KB
- loop/reconciliation-loop.md 3.6 KB
- manifest.json 1.6 KB
- rules/00-golden-invariant.md 1023 B
- rules/01-modality.md 971 B
- rules/02-provenance.md 1012 B
- rules/03-proposal-is-a-lead.md 1.3 KB
- rules/04-graph-freshness.md 2.8 KB
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.
- 2d ago First seen · 44 lines · 44 tokens per session scan A 0e276c79fd11
uigraph is a skill published in the GitHub repository kanetran29/uigraph (6 stars, last pushed 1mo ago), licensed MIT. It adds 44 tokens to every session and 619 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
reference-signal-forms
Explains the mental model and architecture of the code under packages/forms/signals. You MUST use this skill any time you plan to work with code in packages/forms/signals.
xcodebuildmcp-structured-output-review
Use when reviewing XcodeBuildMCP structured output schema changes, schema versioning, manifest outputSchema metadata, and JSON fixture compatibility.
opik-diagnose
Surface the Opik traces worth a developer's attention, ranked by signal — errors, failed tool calls, latency, regressions, and low online-eval scores — plus Diagnostics issues. Reads live/production traces via the SDK (searchtraces and agentinsights) and works with no MCP; uses the MCP issue entity when connected.…
agent-upkeep
Perform one small, scoped maintenance improvement to the Elements monorepo and open a single reviewable pull request. Use this skill for scheduled or unattended upkeep runs that improve unit test coverage for one file, fix one behavioral bug in one module, or move one off ESLint rule toward enforcement to reduce…
cortex-automate
Set up automation — prospective memory triggers, neuro-symbolic rules, and CLAUDE.md sync. Use when the user says 'remind me when', 'trigger when', 'create a rule', 'auto-remember', 'sync to CLAUDE.md', 'push insights', 'set up trigger', 'when I open this file', 'when this keyword appears', or when you want to…
a11y-page-checks
Page-level WCAG checks for full-page or route audits: heading hierarchy, page title uniqueness, focus on route change, and axe test coverage.