CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

add-iron-proxy

Install or refresh Iron Proxy and its official Iron Control web console for NanoClaw. Use when selecting the Iron gateway, adding its management UI, configuring credentials and grants, or restoring the proxy, approval bridge, and agent guidance.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/add-iron-proxy/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, validation-rich skill whose commands are copy-paste ready and whose troubleshooting covers real failure modes. Its weaknesses are token efficiency and structure: large amounts of reference-grade detail are inlined in SKILL.md instead of being split into bundled reference files, and a few referenced artifacts are missing from the bundle.

Suggestions

Move the deep operational detail — the NANOCLAW_GATEWAY_UNCREDENTIALED_READS grant-caching/race semantics and the approval-front architecture description — into a bundled reference file (alongside references/control-plane.md), keeping a one-to-two-line summary and a clear pointer in SKILL.md.

Split the Codex and OpenCode authentication sections into a reference file (or a dedicated auth.md), keeping only the provider-auth entry-point commands and the key warnings inline.

Fix the bundle's dangling references: ship the missing REMOVE.md and versions.json referenced by the body, and move the "Provider credentials on shared model hosts" subsection out from under "Remove" into its own clearly labeled upgrade/refresh section.

DimensionReasoningScore

Conciseness

The body is dense with non-inferable operational detail (timeouts, fail-closed rules, env var defaults), so it avoids explaining things Claude already knows, but sections like the ten-line NANOCLAW_GATEWAY_UNCREDENTIALED_READS grant-cache/race paragraph and the twice-described front architecture (in the install section and again in Validate) are padded beyond overview level. It fits anchor 3 ("could be tightened") better than anchor 4's "minor instances".

3 / 5

Actionability

Guidance is overwhelmingly executable: exact nc:copy path maps, pinned dependency versions, copy-paste tsx commands (setup --with-control, control.ts grant, --allow-host, provider-auth), and env-var defaults with remediation commands. It falls short of anchor 5 only because the Remove workflow directs the reader to a REMOVE.md that is not present in the bundle.

4 / 5

Workflow Clarity

The install→register→dependencies→setup→validate sequence is clear, with an explicit Validate section (build plus a full named test suite) and a Troubleshooting section with diagnosis and recovery steps ("On failure, fix the reported access or service issue and rerun setup"). It misses anchor 5 because the refresh/upgrade procedure is buried in the "Provider credentials on shared model hosts" subsection under Remove, and sequencing across the multiple workflows is only implicit.

4 / 5

Progressive Disclosure

Section headers are good and the one bundled reference is clearly signaled ("Read [references/control-plane.md]... for the exact supported UI, source links, policy example, and verification steps"), but ~19KB of reference-grade detail — the OpenCode authentication minutiae and the credential-rule caching semantics — is inlined rather than split out, and three referenced paths (REMOVE.md, versions.json, docs/gateway-seam.md) are absent from the bundle. This fits anchor 3 (content that should be separate is inline) better than anchor 4's "minor organization gaps".

3 / 5

Total

14

/

20

Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: third-person, concrete, with an explicit "Use when..." trigger clause and product-anchored distinctiveness. Its only weakness is that the action list does not surface the agent-authentication and local-model workflows the body actually covers.

DimensionReasoningScore

Specificity

"Install or refresh Iron Proxy and its official Iron Control web console" plus "selecting the Iron gateway, adding its management UI, configuring credentials and grants, or restoring the proxy, approval bridge, and agent guidance" lists several concrete actions in third person. It stops short of anchor 5 because agent-authentication and local-model workflows covered in the body are absent from the action list.

4 / 5

Completeness

The what is explicit ("Install or refresh Iron Proxy and its official Iron Control web console for NanoClaw") and the when is an explicit "Use when..." clause with concrete trigger activities. Both are clearly and concretely answered, matching anchor 5; the weaker-when phrasing of anchor 4 does not apply.

5 / 5

Trigger Term Quality

Natural product keywords are well covered: "Iron Proxy", "Iron Control", "Iron gateway", "management UI", "credentials", "grants" — terms a user needing this skill would actually say. A few natural phrasings (e.g. console/proxy setup, agent login) are missing, so it fits anchor 4 rather than the comprehensive synonym coverage of anchor 5.

4 / 5

Distinctiveness Conflict Risk

The description is anchored to named products ("Iron Proxy", "Iron Control", "NanoClaw") with distinct triggers like "selecting the Iron gateway", giving it a clear niche with minimal overlap risk against generic gateway or credential skills. This matches anchor 5.

5 / 5

Total

18

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

nanocoai/nanoclaw

Path

.claude/skills/add-iron-proxy/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.