hermes-skill

Name: hermes-skill
Author: AlexAI-MCP

132stars24forks

How to install hermes-skill?

alexai-mcp/hermes-ccc/hermes-skill

$npx -y skills add AlexAI-MCP/hermes-CCC --skill hermes-skill

Installs into the current project.

Summary

Create, improve, inventory, and audit Claude Code skills. Use when turning a solved workflow into a reusable skill, expanding an existing skill, checking trigger quality, or standardizing a skill pack.

Files · 1

View on GitHub

SKILL.md

1	# Hermes Skill
2
3	## Purpose
4
5	- Convert repeated workflows into reusable Claude Code skills.
6	- Keep skills operational rather than descriptive.
7	- Improve trigger quality so the right skill activates at the right time.
8	- Audit skill packs for overlap, drift, and incomplete instructions.
9	- Preserve working patterns after a successful task.
10
11	## Required Skill Anatomy
12
13	- one folder per skill
14	- one `SKILL.md` file per folder
15	- YAML frontmatter at the top
16	- Markdown instructions below the frontmatter
17	- commands, outputs, and failure handling in the body
18
19	## Frontmatter Schema For This Package
20
21	- `name`
22	- `description`
23	- `version`
24	- `author`
25	- `license`
26	- `metadata`
27
28	## Description Rules
29
30	- State what the skill does.
31	- State when to use it.
32	- Mention trigger contexts directly.
33	- Avoid vague phrasing such as "general helper".
34	- Include important boundaries when the skill is domain-specific.
35
36	## Naming Rules
37
38	- use lowercase letters, digits, and hyphens only
39	- keep names short and concrete
40	- prefer operational nouns or verb-noun pairs
41	- avoid project-specific private jargon unless the package is private
42	- do not rename a stable public skill casually
43
44	## Create Workflow
45
46	1. Identify a repeated workflow that was just solved.
47	2. Extract the reusable steps, not the chat narrative.
48	3. List the exact tools, commands, and files involved.
49	4. Identify preconditions and failure modes.
50	5. Write the frontmatter with strong trigger language.
51	6. Write the body as an operator playbook.
52	7. Include output contracts where consistency matters.
53	8. Include examples only when they sharpen execution.
54
55	## Improve Workflow
56
57	1. Read the full skill.
58	2. Ask what failed or felt ambiguous in actual usage.
59	3. Tighten the description if triggering was weak.
60	4. Add missing commands or environment assumptions.
61	5. Add decision rules where the old version left too much ambiguity.
62	6. Add failure handling if the skill assumed a happy path.
63	7. Remove stale instructions and dead commands.
64
65	## Audit Dimensions
66
67	- trigger clarity
68	- instruction quality
69	- command accuracy
70	- line-count sufficiency
71	- duplication with nearby skills
72	- presence of decision rules
73	- presence of output contracts
74	- presence of verification steps
75
76	## What Makes A Good Skill
77
78	- It reduces repeated thinking.
79	- It includes exact actions for non-obvious steps.
80	- It names the right tools and commands.
81	- It narrows ambiguity without overfitting one repo.
82	- It handles the common failure path.
83	- It is faster to use than rediscovering the workflow.
84
85	## What Makes A Bad Skill
86
87	- It repeats generic advice the base model already knows.
88	- It has weak frontmatter and never triggers.
89	- It explains philosophy but omits execution.
90	- It contains stale paths or commands.
91	- It is so narrow that it only matches one past task.
92	- It has no failure modes or verification guidance.
93
94	## Skill Extraction Questions
95
96	- What exact problem was solved?
97	- What sequence of actions worked?
98	- What tools or connectors were essential?
99	- What surprised us during execution?
100	- What constraints would another agent need to know up front?
101	- What output structure made the result useful?
102
103	## Standard Sections To Include
104
105	- purpose
106	- activation signals
107	- inputs
108	- preflight
109	- step-by-step procedure
110	- decision rules
111	- output contract
112	- failure modes
113	- examples
114	- checklist
115
116	## Inventory Workflow
117
118	1. List all skill folders.
119	2. Verify that each folder contains `SKILL.md`.
120	3. Read frontmatter fields.
121	4. Produce a compact table of name, description, and health.
122	5. Flag missing or malformed skills.
123
124	## Overlap Rules
125
126	- Two skills may overlap in domain but should differ in operating goal.
127	- Merge skills if their triggers, commands, and outputs are nearly identical.
128	- Keep separate skills if one is strategic and the other is procedural.
129	- Prefer smaller focused skills over one giant omnibus skill.
130
131	## Updating From Real Usage
132
133	- Capture friction immediately after using a skill.
134	- Update the skill while the failure mode is fresh.
135	- Preserve examples that encode real mistakes.
136	- Replace speculative guidance with validated instructions.
137
138	## Example Improvement Targets
139
140	- add missing prereq checks
141	- tighten "when to use" language
142	- replace obsolete CLI flags
143	- add a markdown response template
144	- split one overloaded skill into two
145
146	## Audit Output Template
147
148	```markdown
149	Skill audit report:
150	- systematic-debugging: Healthy - clear phases, explicit outputs, and verification loop
151	- old-review-skill: Outdated - stale GitHub CLI commands and no failure handling
152	- d

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

hermes-skill

132stars24forks

View on GitHub

How to install hermes-skill?

alexai-mcp/hermes-ccc/hermes-skill

$npx -y skills add AlexAI-MCP/hermes-CCC --skill hermes-skill

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Hermes Skill
2
3	## Purpose
4
5	- Convert repeated workflows into reusable Claude Code skills.
6	- Keep skills operational rather than descriptive.
7	- Improve trigger quality so the right skill activates at the right time.
8	- Audit skill packs for overlap, drift, and incomplete instructions.
9	- Preserve working patterns after a successful task.
10
11	## Required Skill Anatomy
12
13	- one folder per skill
14	- one `SKILL.md` file per folder
15	- YAML frontmatter at the top
16	- Markdown instructions below the frontmatter
17	- commands, outputs, and failure handling in the body
18
19	## Frontmatter Schema For This Package
20
21	- `name`
22	- `description`
23	- `version`
24	- `author`
25	- `license`
26	- `metadata`
27
28	## Description Rules
29
30	- State what the skill does.
31	- State when to use it.
32	- Mention trigger contexts directly.
33	- Avoid vague phrasing such as "general helper".
34	- Include important boundaries when the skill is domain-specific.
35
36	## Naming Rules
37
38	- use lowercase letters, digits, and hyphens only
39	- keep names short and concrete
40	- prefer operational nouns or verb-noun pairs
41	- avoid project-specific private jargon unless the package is private
42	- do not rename a stable public skill casually
43
44	## Create Workflow
45
46	1. Identify a repeated workflow that was just solved.
47	2. Extract the reusable steps, not the chat narrative.
48	3. List the exact tools, commands, and files involved.
49	4. Identify preconditions and failure modes.
50	5. Write the frontmatter with strong trigger language.
51	6. Write the body as an operator playbook.
52	7. Include output contracts where consistency matters.
53	8. Include examples only when they sharpen execution.
54
55	## Improve Workflow
56
57	1. Read the full skill.
58	2. Ask what failed or felt ambiguous in actual usage.
59	3. Tighten the description if triggering was weak.
60	4. Add missing commands or environment assumptions.
61	5. Add decision rules where the old version left too much ambiguity.
62	6. Add failure handling if the skill assumed a happy path.
63	7. Remove stale instructions and dead commands.
64
65	## Audit Dimensions
66
67	- trigger clarity
68	- instruction quality
69	- command accuracy
70	- line-count sufficiency
71	- duplication with nearby skills
72	- presence of decision rules
73	- presence of output contracts
74	- presence of verification steps
75
76	## What Makes A Good Skill
77
78	- It reduces repeated thinking.
79	- It includes exact actions for non-obvious steps.
80	- It names the right tools and commands.
81	- It narrows ambiguity without overfitting one repo.
82	- It handles the common failure path.
83	- It is faster to use than rediscovering the workflow.
84
85	## What Makes A Bad Skill
86
87	- It repeats generic advice the base model already knows.
88	- It has weak frontmatter and never triggers.
89	- It explains philosophy but omits execution.
90	- It contains stale paths or commands.
91	- It is so narrow that it only matches one past task.
92	- It has no failure modes or verification guidance.
93
94	## Skill Extraction Questions
95
96	- What exact problem was solved?
97	- What sequence of actions worked?
98	- What tools or connectors were essential?
99	- What surprised us during execution?
100	- What constraints would another agent need to know up front?
101	- What output structure made the result useful?
102
103	## Standard Sections To Include
104
105	- purpose
106	- activation signals
107	- inputs
108	- preflight
109	- step-by-step procedure
110	- decision rules
111	- output contract
112	- failure modes
113	- examples
114	- checklist
115
116	## Inventory Workflow
117
118	1. List all skill folders.
119	2. Verify that each folder contains `SKILL.md`.
120	3. Read frontmatter fields.
121	4. Produce a compact table of name, description, and health.
122	5. Flag missing or malformed skills.
123
124	## Overlap Rules
125
126	- Two skills may overlap in domain but should differ in operating goal.
127	- Merge skills if their triggers, commands, and outputs are nearly identical.
128	- Keep separate skills if one is strategic and the other is procedural.
129	- Prefer smaller focused skills over one giant omnibus skill.
130
131	## Updating From Real Usage
132
133	- Capture friction immediately after using a skill.
134	- Update the skill while the failure mode is fresh.
135	- Preserve examples that encode real mistakes.
136	- Replace speculative guidance with validated instructions.
137
138	## Example Improvement Targets
139
140	- add missing prereq checks
141	- tighten "when to use" language
142	- replace obsolete CLI flags
143	- add a markdown response template
144	- split one overloaded skill into two
145
146	## Audit Output Template
147
148	```markdown
149	Skill audit report:
150	- systematic-debugging: Healthy - clear phases, explicit outputs, and verification loop
151	- old-review-skill: Outdated - stale GitHub CLI commands and no failure handling
152	- d

People who liked this also viewed

6 picks

Type