skill-debrief

Name: skill-debrief
Author: computerlovetech

448stars36forks

How to install skill-debrief?

computerlovetech/agr/skill-debrief

$npx -y skills add computerlovetech/agr --skill skill-debrief

Installs into the current project.

Summary

Debrief an AI agent skill (SKILL.md) after using it — capture session feedback or a retrospective and fold it back into the skill. Use whenever the user says: "debrief the X skill", "let's debrief X", "retrospective on X", "feedback on X skill", "improve the X skill", "update the

Files · 1

View on GitHub

SKILL.md

1	# Skill Debrief
2
3	Capture lessons from a session into the skill that drove it. The default
4	shape is listen → propose → align → apply → re-install.
5
6	## When to use
7
8	Trigger when the user wants to debrief an existing SKILL.md based on what
9	happened in the session. Examples:
10
11	- "debrief the X skill" / "let's debrief X"
12	- "retrospective on X" / "feedback on X"
13	- "improve the X skill" / "let's update X based on what we learned"
14	- "X skill should also handle …"
15	- "X didn't trigger when it should have"
16
17	Do NOT use this skill for:
18
19	- Greenfield skill authoring. Use `agr init` to scaffold a SKILL.md and
20	defer the body content to the user — or to a dedicated authoring skill
21	such as `anthropics/skills/skill-creator`
22	(`agr add anthropics/skills/skill-creator`).
23	- Installing / syncing / removing skills. That's plain `agr` CLI work
24	(`agr add`, `agr sync`, `agr upgrade`, `agr remove`).
25
26	## Step 1: Identify the skill
27
28	Ask which skill is being improved if it isn't obvious from context. Then
29	locate the source:
30
31	```bash
32	agr list # see installed deps and short names
33	ls skills/ # in-repo source if present
34	cat agr.toml # see whether the dep is local-path or remote
35	```
36
37	Two cases — they have different update paths:
38
39	\| Case \| Source location \| Update path \|
40	\|---\|---\|---\|
41	\| In-repo (`{path = "./skills/<name>", type = "skill"}` in `agr.toml`) \| `skills/<name>/` \| Edit source → commit → `agr upgrade <name>` \|
42	\| Remote (`{handle = "user/repo/<name>", …}`) \| Upstream GitHub repo \| Cannot edit directly — see Step 5 \|
43
44	If the skill isn't installed at all but the user wants to improve it, ask
45	whether to add it first (and which case applies).
46
47	## Step 2: Receive feedback
48
49	Listen. The user invoked this skill because they have something to say
50	— let them say it. Do not interrogate. Do not run a checklist of questions
51	at them. Take in whatever they offer, in whatever shape they offer it.
52
53	Only ask a clarifying question if you genuinely cannot proceed without one
54	(e.g. the user named a skill that doesn't exist, or two skills share the
55	name and you need to disambiguate). Even then, ask the minimum.
56
57	Be dynamic. The user may surface things in any shape — a single
58	sentence ("the description should also fire on X"), a structured list, or a
59	ramble that you need to distill. They may also surface things outside the
60	standard buckets below (rename a section, restructure `references/`, change
61	output format, drop a deprecated workflow, fix a typo). Apply whatever the
62	user actually says.
63
64	The buckets below are a mental map for you when distilling what you
65	heard, not a checklist to recite at the user:
66
67	- `description` / triggers — under-fired or over-fired
68	- Gotchas / boundaries — a foot-gun the skill didn't warn about
69	- Workflow steps — missing, wrong, or out of order
70	- References — a topic kept needing more depth → new `references/<topic>.md`
71	- Examples / output format — vague where it should be concrete
72	- Pruning — outdated content that misled
73
74	## Step 3: Propose changes
75
76	Summarize what you heard, then propose specific edits. Format:
77
78	> Proposed changes to `skills/<name>/SKILL.md` (and any references):
79	>
80	> 1. Description — add trigger phrase "…" (because: …)
81	> 2. Boundaries — add: never X (because: discovered this in session)
82	> 3. New section "Y" — describes the workflow that was missing
83	>
84	> Want me to apply these, revise, or add more?
85
86	For small edits, show the exact diff inline. For larger changes, summarize
87	first and apply section by section.
88
89	Wait for explicit user approval before editing. "yes" / "go ahead" /
90	similar. If the user revises, loop back to Step 2 or 3.
91
92	## Step 4: Apply (in-repo case)
93
94	Edit the source file(s) under `skills/<name>/`. Then:
95
96	```bash
97	git status # confirm only the intended files changed
98	git add skills/<name>/
99	git commit -m "skill(<name>): <one-line summary>"
100	agr upgrade <name> # re-installs into all configured tools, refreshes a

People who liked this also viewed

6 picks

Type

skill-debrief

448stars36forks

View on GitHub

How to install skill-debrief?

computerlovetech/agr/skill-debrief

$npx -y skills add computerlovetech/agr --skill skill-debrief

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Skill Debrief
2
3	Capture lessons from a session into the skill that drove it. The default
4	shape is listen → propose → align → apply → re-install.
5
6	## When to use
7
8	Trigger when the user wants to debrief an existing SKILL.md based on what
9	happened in the session. Examples:
10
11	- "debrief the X skill" / "let's debrief X"
12	- "retrospective on X" / "feedback on X"
13	- "improve the X skill" / "let's update X based on what we learned"
14	- "X skill should also handle …"
15	- "X didn't trigger when it should have"
16
17	Do NOT use this skill for:
18
19	- Greenfield skill authoring. Use `agr init` to scaffold a SKILL.md and
20	defer the body content to the user — or to a dedicated authoring skill
21	such as `anthropics/skills/skill-creator`
22	(`agr add anthropics/skills/skill-creator`).
23	- Installing / syncing / removing skills. That's plain `agr` CLI work
24	(`agr add`, `agr sync`, `agr upgrade`, `agr remove`).
25
26	## Step 1: Identify the skill
27
28	Ask which skill is being improved if it isn't obvious from context. Then
29	locate the source:
30
31	```bash
32	agr list # see installed deps and short names
33	ls skills/ # in-repo source if present
34	cat agr.toml # see whether the dep is local-path or remote
35	```
36
37	Two cases — they have different update paths:
38
39	\| Case \| Source location \| Update path \|
40	\|---\|---\|---\|
41	\| In-repo (`{path = "./skills/<name>", type = "skill"}` in `agr.toml`) \| `skills/<name>/` \| Edit source → commit → `agr upgrade <name>` \|
42	\| Remote (`{handle = "user/repo/<name>", …}`) \| Upstream GitHub repo \| Cannot edit directly — see Step 5 \|
43
44	If the skill isn't installed at all but the user wants to improve it, ask
45	whether to add it first (and which case applies).
46
47	## Step 2: Receive feedback
48
49	Listen. The user invoked this skill because they have something to say
50	— let them say it. Do not interrogate. Do not run a checklist of questions
51	at them. Take in whatever they offer, in whatever shape they offer it.
52
53	Only ask a clarifying question if you genuinely cannot proceed without one
54	(e.g. the user named a skill that doesn't exist, or two skills share the
55	name and you need to disambiguate). Even then, ask the minimum.
56
57	Be dynamic. The user may surface things in any shape — a single
58	sentence ("the description should also fire on X"), a structured list, or a
59	ramble that you need to distill. They may also surface things outside the
60	standard buckets below (rename a section, restructure `references/`, change
61	output format, drop a deprecated workflow, fix a typo). Apply whatever the
62	user actually says.
63
64	The buckets below are a mental map for you when distilling what you
65	heard, not a checklist to recite at the user:
66
67	- `description` / triggers — under-fired or over-fired
68	- Gotchas / boundaries — a foot-gun the skill didn't warn about
69	- Workflow steps — missing, wrong, or out of order
70	- References — a topic kept needing more depth → new `references/<topic>.md`
71	- Examples / output format — vague where it should be concrete
72	- Pruning — outdated content that misled
73
74	## Step 3: Propose changes
75
76	Summarize what you heard, then propose specific edits. Format:
77
78	> Proposed changes to `skills/<name>/SKILL.md` (and any references):
79	>
80	> 1. Description — add trigger phrase "…" (because: …)
81	> 2. Boundaries — add: never X (because: discovered this in session)
82	> 3. New section "Y" — describes the workflow that was missing
83	>
84	> Want me to apply these, revise, or add more?
85
86	For small edits, show the exact diff inline. For larger changes, summarize
87	first and apply section by section.
88
89	Wait for explicit user approval before editing. "yes" / "go ahead" /
90	similar. If the user revises, loop back to Step 2 or 3.
91
92	## Step 4: Apply (in-repo case)
93
94	Edit the source file(s) under `skills/<name>/`. Then:
95
96	```bash
97	git status # confirm only the intended files changed
98	git add skills/<name>/
99	git commit -m "skill(<name>): <one-line summary>"
100	agr upgrade <name> # re-installs into all configured tools, refreshes a

People who liked this also viewed

6 picks

Type