skill-update

Name: skill-update
Author: transilienceai

423stars74forks

How to install skill-update?

transilienceai/communitytools/skill-update

$npx -y skills add transilienceai/communitytools --skill skill-update

Installs into the current project.

Summary

Skill creation, update and management — generates skill directory structure, validates against best practices, enforces line count limits. Use when creating, updating, or improving skills.

Files · 1

View on GitHub

SKILL.md

1	# Skill Update
2
3	Generate or refine Claude Code skills following Anthropic best practices.
4
5	## Hard caps (enforced by `scripts/skill_linter.py`)
6
7	- `SKILL.md` ≤ 150 lines
8	- `reference/.md` ≤ 200 lines (`reference/scenarios/.md` ≤ 400 lines)
9	- `README.md` ≤ 100 lines
10	- Every `SKILL.md` has YAML frontmatter with `name` + `description`
11	- No `DO NOT` / `MUST NOT` / `NEVER` outside an `## Anti-Patterns` section
12	- No challenge-specific identifiers (machine names, lab IDs, lab IPs, preserved flags)
13	- Every Markdown link resolves to an existing file
14	- Every reference file is linked from at least one other file (no orphans)
15
16	## Principles
17
18	- Brevity first. Every file short, simple, human-readable. Challenge every token.
19	- Progressive disclosure. SKILL.md navigates; `reference/` holds detail; `reference/scenarios/` holds concrete exploit flows.
20	- Separation of concern. SKILL.md = WHAT + when. `reference/role-*.md` = HOW agents behave when spawned.
21	- Single canonical home for any cross-cutting rule (output discipline, credential loading, brute-force, etc.). Other files reference, never restate.
22
23	## File structure
24
25	```
26	skills/<skill-name>/
27	├── SKILL.md # ≤150 lines, YAML + navigation
28	├── reference/
29	│ ├── *-principles.md # ≤150 lines (decision tree)
30	│ ├── INDEX.md
31	│ ├── *.md # patterns, ≤200 lines
32	│ └── scenarios/
33	│ └── <category>/
34	│ └── *.md # ≤400 lines, self-contained
35	└── README.md # optional, ≤100 lines
36	```
37
38	## SKILL.md template
39
40	```yaml
41	---
42	name: <skill-name>
43	description: What it does AND when to use. Include trigger phrases.
44	---
45
46	# <Skill Name>
47
48	<one-paragraph scope>
49
50	## When to use
51
52	- <bullet>
53
54	## Workflow / Quick start
55
56	<≤30 lines>
57
58	## References
59
60	- [reference/...](reference/...)
61
62	## Anti-Patterns
63
64	- <when negative framing is genuinely needed, put it here>
65	```
66
67	## When to update an existing skill
68
69	Process the techniques and failure modes from completed engagements. Promote a learning to the skill base only if all four hold:
70
71	1. Generalizable. Reusable pattern, not target-specific lore. No machine names, lab IDs, target IPs, preserved flags, writeup attributions.
72	2. Material improvement. Adds coverage, efficiency, or decision-quality for future engagements.
73	3. Not already captured elsewhere in the skill base. (`scripts/skill_linter.py` flags duplicates.)
74	4. Minimal footprint. Prefer extending an existing entry over adding a new file. Keep the base lean and high-signal.
75
76	## Reframing recipe
77
78	Always frame as a reusable pattern: "when encountering X condition, try Y approach" — never "on box-N, Y worked". Use `<TARGET_IP>`, `<DC_FQDN>`, `<DOMAIN>` placeholders in tool examples.
79
80	## Pre-write check
81
82	Before writing, run `python3 scripts/skill_linter.py`. Reject any change that:
83	- Re-introduces challenge-specific lore.
84	- Pushes a `SKILL.md` past 150 or a reference past its cap.
85	- Duplicates a single-owner rule (brute-force, output discipline, env-reader).
86	- Adds `DO NOT` / `MUST NOT` / `NEVER` outside an Anti-Patterns block.
87
88	## Output
89
90	Concise change report:
91	- Updated. File + one-line summary of edit.
92	- Skipped. Notable findings intentionally not added, with brief reasoning.
93	- No changes. State explicitly when nothing warranted an update.
94
95	## Reference
96
97	- [STRUCTURE.md](reference/STRUCTURE.md) — directory layout requirements.
98	- [FRONTMATTER.md](reference/FRONTMATTER.md) — YAML rules.
99	- [CONTENT.md](reference/CONTENT.md) — writing guidelines.
100
101	## Anti-Patterns
102
103	- Creating CHANGELOG.md / SUMMARY.md / VERIFICATION.md auxiliary files.
104	- Meta-documentation about the creation process inside the skill itself.
105	- Verbose inline templates and examples (link to `reference/` instead).
106	- Re-introducing duplicate rule prose (brute-force, output-dir, env-reader).
107	- Files past their cap — split into `reference/` immediately.

People who liked this also viewed

6 picks

Type

skill-update

423stars74forks

View on GitHub

How to install skill-update?

transilienceai/communitytools/skill-update

$npx -y skills add transilienceai/communitytools --skill skill-update

Installs into the current project.

Summary

Skill creation, update and management — generates skill directory structure, validates against best practices, enforces line count limits. Use when creating, updating, or improving skills.

Files · 1

View on GitHub

SKILL.md

1	# Skill Update
2
3	Generate or refine Claude Code skills following Anthropic best practices.
4
5	## Hard caps (enforced by `scripts/skill_linter.py`)
6
7	- `SKILL.md` ≤ 150 lines
8	- `reference/.md` ≤ 200 lines (`reference/scenarios/.md` ≤ 400 lines)
9	- `README.md` ≤ 100 lines
10	- Every `SKILL.md` has YAML frontmatter with `name` + `description`
11	- No `DO NOT` / `MUST NOT` / `NEVER` outside an `## Anti-Patterns` section
12	- No challenge-specific identifiers (machine names, lab IDs, lab IPs, preserved flags)
13	- Every Markdown link resolves to an existing file
14	- Every reference file is linked from at least one other file (no orphans)
15
16	## Principles
17
18	- Brevity first. Every file short, simple, human-readable. Challenge every token.
19	- Progressive disclosure. SKILL.md navigates; `reference/` holds detail; `reference/scenarios/` holds concrete exploit flows.
20	- Separation of concern. SKILL.md = WHAT + when. `reference/role-*.md` = HOW agents behave when spawned.
21	- Single canonical home for any cross-cutting rule (output discipline, credential loading, brute-force, etc.). Other files reference, never restate.
22
23	## File structure
24
25	```
26	skills/<skill-name>/
27	├── SKILL.md # ≤150 lines, YAML + navigation
28	├── reference/
29	│ ├── *-principles.md # ≤150 lines (decision tree)
30	│ ├── INDEX.md
31	│ ├── *.md # patterns, ≤200 lines
32	│ └── scenarios/
33	│ └── <category>/
34	│ └── *.md # ≤400 lines, self-contained
35	└── README.md # optional, ≤100 lines
36	```
37
38	## SKILL.md template
39
40	```yaml
41	---
42	name: <skill-name>
43	description: What it does AND when to use. Include trigger phrases.
44	---
45
46	# <Skill Name>
47
48	<one-paragraph scope>
49
50	## When to use
51
52	- <bullet>
53
54	## Workflow / Quick start
55
56	<≤30 lines>
57
58	## References
59
60	- [reference/...](reference/...)
61
62	## Anti-Patterns
63
64	- <when negative framing is genuinely needed, put it here>
65	```
66
67	## When to update an existing skill
68
69	Process the techniques and failure modes from completed engagements. Promote a learning to the skill base only if all four hold:
70
71	1. Generalizable. Reusable pattern, not target-specific lore. No machine names, lab IDs, target IPs, preserved flags, writeup attributions.
72	2. Material improvement. Adds coverage, efficiency, or decision-quality for future engagements.
73	3. Not already captured elsewhere in the skill base. (`scripts/skill_linter.py` flags duplicates.)
74	4. Minimal footprint. Prefer extending an existing entry over adding a new file. Keep the base lean and high-signal.
75
76	## Reframing recipe
77
78	Always frame as a reusable pattern: "when encountering X condition, try Y approach" — never "on box-N, Y worked". Use `<TARGET_IP>`, `<DC_FQDN>`, `<DOMAIN>` placeholders in tool examples.
79
80	## Pre-write check
81
82	Before writing, run `python3 scripts/skill_linter.py`. Reject any change that:
83	- Re-introduces challenge-specific lore.
84	- Pushes a `SKILL.md` past 150 or a reference past its cap.
85	- Duplicates a single-owner rule (brute-force, output discipline, env-reader).
86	- Adds `DO NOT` / `MUST NOT` / `NEVER` outside an Anti-Patterns block.
87
88	## Output
89
90	Concise change report:
91	- Updated. File + one-line summary of edit.
92	- Skipped. Notable findings intentionally not added, with brief reasoning.
93	- No changes. State explicitly when nothing warranted an update.
94
95	## Reference
96
97	- [STRUCTURE.md](reference/STRUCTURE.md) — directory layout requirements.
98	- [FRONTMATTER.md](reference/FRONTMATTER.md) — YAML rules.
99	- [CONTENT.md](reference/CONTENT.md) — writing guidelines.
100
101	## Anti-Patterns
102
103	- Creating CHANGELOG.md / SUMMARY.md / VERIFICATION.md auxiliary files.
104	- Meta-documentation about the creation process inside the skill itself.
105	- Verbose inline templates and examples (link to `reference/` instead).
106	- Re-introducing duplicate rule prose (brute-force, output-dir, env-reader).
107	- Files past their cap — split into `reference/` immediately.

People who liked this also viewed

6 picks

Type