agent-ready

Name: agent-ready
Author: redhat-developer

15stars31forks

How to install agent-ready?

redhat-developer/rhdh-skill/agent-ready

$npx -y skills add redhat-developer/rhdh-skill --skill agent-ready

Installs into the current project.

Summary

Assesses a git repository's readiness for use by AI coding agents using the agentready CLI, then walks through and addresses each gap. RHDH-aware: detects RHDH repositories and uses rhdh-repos.md context to pre-fill AGENTS.md and skip inapplicable findings. Use when asked to "ass

Files · 1

View on GitHub

SKILL.md

1	## Prerequisites
2
3	`uvx` is a hard dependency. Verify it is available before any other step:
4
5	```bash
6	uvx --version
7	```
8
9	If missing, stop: "`uvx` is required. Install via `pip install uv` or see [uv installation](https://docs.astral.sh/uv/getting-started/installation/)."
10
11	## Step 1: Mode selection
12
13	If no path was provided, present a structured choice:
14
15	- Single repo — assess the current working directory (default)
16	- Batch — assess all RHDH repositories (see Batch mode below)
17
18	If a path was provided, skip this and proceed to Step 2.
19
20	## Step 2: Setup
21
22	Path: Use the provided path, or `.` for the current directory. Validate it is a git repository:
23
24	```bash
25	git -C . rev-parse --is-inside-work-tree # replace . with path if provided
26	```
27
28	If not a git repository, stop and tell the user.
29
30	RHDH detection: Check the repo's git remote URL:
31
32	```bash
33	git -C <path> remote get-url origin 2>/dev/null
34	```
35
36	Attempt to read `~/.claude/skills/rhdh/references/rhdh-repos.md`. If the file does not exist, skip RHDH detection and proceed with generic assessment — do not stop or warn the user. If found, check whether the remote URL matches any repo's upstream URL. If matched, note the repo name, tech stack, key paths, and conventions — these inform AGENTS.md generation and finding triage. Store as `rhdh_context`.
37
38	Config file: Only use a config file if the user explicitly provided one. Do not ask.
39
40	## Step 3: Run the assessment
41
42	```bash
43	REPORT_DIR=$(mktemp -d) # on Windows: use %TEMP% or Python tempfile
44	uvx --from git+https://github.com/ambient-code/agentready agentready -- assess \
45	-o "$REPORT_DIR" \
46	<path>
47	```
48
49	Append `-c <config-path>` if the user provided a config file.
50
51	Note the value of `$REPORT_DIR` — shell variables do not persist across tool calls.
52
53	Parse `$REPORT_DIR/assessment-latest.json`. Extract:
54	- `overall_score`, `certification_level`
55	- `findings` — each with `attribute.id`, `attribute.tier`, `attribute.default_weight`, `attribute.name`, `status`, `score`, `evidence`, `remediation`
56
57	## Step 4: Present summary
58
59	```
60	Score: <overall_score>/100 — <certification_level>
61	Failing: <N> findings (<N1> Tier 1, <N2> Tier 2, ...)
62	```
63
64	If no failing findings, congratulate the user and stop.
65
66	Otherwise ask:
67
68	> "Fix applicable findings automatically, or review each one individually?
69	> auto (default) — apply self-contained fixes immediately; prompt only when input is needed
70	> review — prompt yes/skip/defer/quit for every finding"
71
72	Default to auto if the user says yes, presses Enter, or says "fix everything".
73
74	## Step 5: Work through findings
75
76	Work only through `status == "fail"` findings. Skip `not_applicable` and `pass` silently.
77
78	Sort order: ascending tier, then descending `attribute.default_weight` within each tier.
79
80	If `rhdh_context` is set, skip findings that clearly don't apply to the detected tech stack (e.g., lock file checks for a Bash-only repo, `src/` layout for a GitOps YAML repo) — note them in the summary.
81
82	### Auto mode
83
84	Apply each fix without prompting unless:
85
86	- The fix requires project-specific input (CI platform, package ecosystem)
87	- The finding might not apply to this repo type — present it and ask whether to apply or skip
88
89	Skip without prompting: ADRs, design intent, architecture decisions — these require human rationale. Note them in the final summary.
90
91	For the `agent_instructions` finding, follow the inline AGENTS.md generation in the `agent_instructions` section below — this applies in both auto and review modes.
92
93	After processing, list what was applied, prompted, and skipped, then proceed to Step 6.
94
95	### Review mode
96
97	For each finding:
98
99	```
100	[Tier <N>] <attribute.name> — <score>/100
101	Evidence: <evidence items>
102
103	Remediation: <remediation.summary>
104
105	Apply this fix? [yes / skip / defer / quit]
106	```
107
108	yes — apply the fix, then move to the next finding.
109	skip — move on; do not revisit. Use this if the finding doesn't apply to this repo.
110	defer — note it; surface again after re-run.
111	quit — stop immediately.
112
113	ADR and design intent findings: Do not use JSON remediation. Ask instead:
114
115	> "Do you have any architectural decisions worth capturing? Describe the decision and rationale — I'll write the ADR. Skip to add manually later."
116
117	Write only if the user provides input. Never invent rationale.
118
119	### `agent_instructions` finding (both modes)
120
121	Generate `AGENTS.md` and `CLAUDE.md` inline — do not delegate to another skill.

People who liked this also viewed

6 picks

Type

agent-ready

15stars31forks

View on GitHub

How to install agent-ready?

redhat-developer/rhdh-skill/agent-ready

$npx -y skills add redhat-developer/rhdh-skill --skill agent-ready

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	## Prerequisites
2
3	`uvx` is a hard dependency. Verify it is available before any other step:
4
5	```bash
6	uvx --version
7	```
8
9	If missing, stop: "`uvx` is required. Install via `pip install uv` or see [uv installation](https://docs.astral.sh/uv/getting-started/installation/)."
10
11	## Step 1: Mode selection
12
13	If no path was provided, present a structured choice:
14
15	- Single repo — assess the current working directory (default)
16	- Batch — assess all RHDH repositories (see Batch mode below)
17
18	If a path was provided, skip this and proceed to Step 2.
19
20	## Step 2: Setup
21
22	Path: Use the provided path, or `.` for the current directory. Validate it is a git repository:
23
24	```bash
25	git -C . rev-parse --is-inside-work-tree # replace . with path if provided
26	```
27
28	If not a git repository, stop and tell the user.
29
30	RHDH detection: Check the repo's git remote URL:
31
32	```bash
33	git -C <path> remote get-url origin 2>/dev/null
34	```
35
36	Attempt to read `~/.claude/skills/rhdh/references/rhdh-repos.md`. If the file does not exist, skip RHDH detection and proceed with generic assessment — do not stop or warn the user. If found, check whether the remote URL matches any repo's upstream URL. If matched, note the repo name, tech stack, key paths, and conventions — these inform AGENTS.md generation and finding triage. Store as `rhdh_context`.
37
38	Config file: Only use a config file if the user explicitly provided one. Do not ask.
39
40	## Step 3: Run the assessment
41
42	```bash
43	REPORT_DIR=$(mktemp -d) # on Windows: use %TEMP% or Python tempfile
44	uvx --from git+https://github.com/ambient-code/agentready agentready -- assess \
45	-o "$REPORT_DIR" \
46	<path>
47	```
48
49	Append `-c <config-path>` if the user provided a config file.
50
51	Note the value of `$REPORT_DIR` — shell variables do not persist across tool calls.
52
53	Parse `$REPORT_DIR/assessment-latest.json`. Extract:
54	- `overall_score`, `certification_level`
55	- `findings` — each with `attribute.id`, `attribute.tier`, `attribute.default_weight`, `attribute.name`, `status`, `score`, `evidence`, `remediation`
56
57	## Step 4: Present summary
58
59	```
60	Score: <overall_score>/100 — <certification_level>
61	Failing: <N> findings (<N1> Tier 1, <N2> Tier 2, ...)
62	```
63
64	If no failing findings, congratulate the user and stop.
65
66	Otherwise ask:
67
68	> "Fix applicable findings automatically, or review each one individually?
69	> auto (default) — apply self-contained fixes immediately; prompt only when input is needed
70	> review — prompt yes/skip/defer/quit for every finding"
71
72	Default to auto if the user says yes, presses Enter, or says "fix everything".
73
74	## Step 5: Work through findings
75
76	Work only through `status == "fail"` findings. Skip `not_applicable` and `pass` silently.
77
78	Sort order: ascending tier, then descending `attribute.default_weight` within each tier.
79
80	If `rhdh_context` is set, skip findings that clearly don't apply to the detected tech stack (e.g., lock file checks for a Bash-only repo, `src/` layout for a GitOps YAML repo) — note them in the summary.
81
82	### Auto mode
83
84	Apply each fix without prompting unless:
85
86	- The fix requires project-specific input (CI platform, package ecosystem)
87	- The finding might not apply to this repo type — present it and ask whether to apply or skip
88
89	Skip without prompting: ADRs, design intent, architecture decisions — these require human rationale. Note them in the final summary.
90
91	For the `agent_instructions` finding, follow the inline AGENTS.md generation in the `agent_instructions` section below — this applies in both auto and review modes.
92
93	After processing, list what was applied, prompted, and skipped, then proceed to Step 6.
94
95	### Review mode
96
97	For each finding:
98
99	```
100	[Tier <N>] <attribute.name> — <score>/100
101	Evidence: <evidence items>
102
103	Remediation: <remediation.summary>
104
105	Apply this fix? [yes / skip / defer / quit]
106	```
107
108	yes — apply the fix, then move to the next finding.
109	skip — move on; do not revisit. Use this if the finding doesn't apply to this repo.
110	defer — note it; surface again after re-run.
111	quit — stop immediately.
112
113	ADR and design intent findings: Do not use JSON remediation. Ask instead:
114
115	> "Do you have any architectural decisions worth capturing? Describe the decision and rationale — I'll write the ADR. Skip to add manually later."
116
117	Write only if the user provides input. Never invent rationale.
118
119	### `agent_instructions` finding (both modes)
120
121	Generate `AGENTS.md` and `CLAUDE.md` inline — do not delegate to another skill.

People who liked this also viewed

6 picks

Type