discover

Name: discover
Author: anombyte93

582stars57forks

How to install discover?

anombyte93/prd-taskmaster/discover

$npx -y skills add anombyte93/prd-taskmaster --skill discover

Installs into the current project.

Summary

Phase 1 of the prd-taskmaster pipeline: brainstorm-driven discovery. Delegates to superpowers:brainstorming in Interactive Mode (one adaptive question at a time), or self-brainstorms in Autonomous Mode when no user is present. Intercepts before the brainstorming chain hands off t

Files · 1

View on GitHub

SKILL.md

1	# Phase 1: Discover
2
3	Declarative phase skill. Invoked by the prd-taskmaster orchestrator when
4	`current_phase` is `DISCOVER`. Never called directly by a user.
5
6	The one rule: **invoke `superpowers:brainstorming` for discovery, intercept
7	before it chains to `writing-plans` — we control the exit, not the brainstorm
8	skill.**
9
10	## Entry gate
11
12	1. Call `mcp__plugin_prd_go__check_gate(phase="DISCOVER", evidence={})` for diagnostics.
13
14	`check_gate` is an EXIT gate: it verifies the evidence to advance, not to enter.
15	On first DISCOVER entry you have no evidence yet (the User Approval / Self-Approval
16	Gate below produces `user_approved=true` OR `auto_classification=CLEAR with
17	assumptions_documented`), so a `gate_passed: false` here is EXPECTED — the state
18	machine's legal transitions already guarantee only legal entry.
19
20	- First entry (no evidence yet): note the result and continue with the Procedure.
21	- Re-entry: if the gate reports violations, report them and stop — it protects
22	against re-running a completed phase or skipping ahead from SETUP.
23	2. Detect execution context. If any of the following signals are present,
24	switch to Autonomous Mode:
25	- `.claude/ralph-loop.local.md` exists in the project root
26	- An `auto-enter` / `auto-approve` daemon is running against this session
27	- The skill was invoked with an explicit `--autonomous` flag
28	- Parent orchestrator is a cron, `/pentest-wtf`, or `/ralph-loop`
29	Otherwise proceed in Interactive Mode (default).
30
31	## Discovery checklist
32
33	Copy into your response before running the procedure:
34
35	```
36	DISCOVERY CHECKLIST:
37	- [ ] Mode detected (Interactive vs Autonomous)
38	- [ ] Goal captured from skill args or soul purpose
39	- [ ] Adaptive questions completed (one at a time)
40	- [ ] Constraints extracted and listed
41	- [ ] Scale classified (Solo / Team / Enterprise)
42	- [ ] Discovery summary captured for GENERATE phase
43	- [ ] User approved (Interactive) or summary committed (Autonomous)
44	```
45
46	## Interactive Mode (default — user present)
47
48	1. Take the user's goal / description from the skill invocation args.
49	2. Invoke `superpowers:brainstorming` with the goal as input.
50	3. Brainstorming runs its adaptive question flow — one domain-agnostic question
51	at a time. Let it drive the Q&A rhythm.
52	4. INTERCEPT POINT: when brainstorming signals readiness to chain to
53	`writing-plans`, STOP. Do NOT let it invoke `writing-plans`. Capture the
54	brainstorm output (design, requirements, decisions) into local state
55	instead. The prd-taskmaster orchestrator owns the handoff — not
56	`superpowers:brainstorming`.
57	5. Present the summary to the user for approval via `AskUserQuestion` (see
58	User Approval Gate below).
59
60	## Autonomous Mode (no user present)
61
62	Do NOT invoke `superpowers:brainstorming` — it blocks on user input and
63	will stall an unattended session. Instead, self-brainstorm using this
64	template:
65
66	1. Read the goal statement from skill args or
67	`session-context/CLAUDE-soul-purpose.md`.
68	2. Read `session-context/CLAUDE-activeContext.md` for project context.
69	3. Write discovery notes directly to
70	`session-context/discovery-{timestamp}.md` answering every question the
71	interactive flow would ask:
72	- Who is this for?
73	- What problem does it solve?
74	- What are the success metrics?
75	- What are the constraints (tech stack, timeline, team, budget,
76	integrations, regulatory)?
77	- What's explicitly out of scope?
78	- What's the scale (Solo / Team / Enterprise)?
79	4. Self-approve: the skill acts as both interrogator and approver. Document
80	assumptions explicitly so the user can audit them on wake-up.
81	5. Commit the discovery file. The git history becomes the audit trail — if
82	the user later disagrees, they can reset to that commit and re-run.
83
84	Autonomous mode is first-class, not degraded. A well-run autonomous
85	discovery produces a spec the user reads on wake-up and says "yes, that's
86	what I meant" without edits. If you find yourself needing to ask more than
87	two questions the user didn't anticipate, the discovery is under-specified —
88	stop and write a handoff note instead of proceeding.
89
90	## User Approval Gate (Interactive Mode)
91
92	After brainstorming completes, present via `AskUserQuestion`:
93
94	```
95	Discovery Complete:
96	Goal: [one sentence]
97	Audience: [who it's for]
98	Approac

People who liked this also viewed

6 picks

Type

discover

582stars57forks

View on GitHub

How to install discover?

anombyte93/prd-taskmaster/discover

$npx -y skills add anombyte93/prd-taskmaster --skill discover

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Phase 1: Discover
2
3	Declarative phase skill. Invoked by the prd-taskmaster orchestrator when
4	`current_phase` is `DISCOVER`. Never called directly by a user.
5
6	The one rule: **invoke `superpowers:brainstorming` for discovery, intercept
7	before it chains to `writing-plans` — we control the exit, not the brainstorm
8	skill.**
9
10	## Entry gate
11
12	1. Call `mcp__plugin_prd_go__check_gate(phase="DISCOVER", evidence={})` for diagnostics.
13
14	`check_gate` is an EXIT gate: it verifies the evidence to advance, not to enter.
15	On first DISCOVER entry you have no evidence yet (the User Approval / Self-Approval
16	Gate below produces `user_approved=true` OR `auto_classification=CLEAR with
17	assumptions_documented`), so a `gate_passed: false` here is EXPECTED — the state
18	machine's legal transitions already guarantee only legal entry.
19
20	- First entry (no evidence yet): note the result and continue with the Procedure.
21	- Re-entry: if the gate reports violations, report them and stop — it protects
22	against re-running a completed phase or skipping ahead from SETUP.
23	2. Detect execution context. If any of the following signals are present,
24	switch to Autonomous Mode:
25	- `.claude/ralph-loop.local.md` exists in the project root
26	- An `auto-enter` / `auto-approve` daemon is running against this session
27	- The skill was invoked with an explicit `--autonomous` flag
28	- Parent orchestrator is a cron, `/pentest-wtf`, or `/ralph-loop`
29	Otherwise proceed in Interactive Mode (default).
30
31	## Discovery checklist
32
33	Copy into your response before running the procedure:
34
35	```
36	DISCOVERY CHECKLIST:
37	- [ ] Mode detected (Interactive vs Autonomous)
38	- [ ] Goal captured from skill args or soul purpose
39	- [ ] Adaptive questions completed (one at a time)
40	- [ ] Constraints extracted and listed
41	- [ ] Scale classified (Solo / Team / Enterprise)
42	- [ ] Discovery summary captured for GENERATE phase
43	- [ ] User approved (Interactive) or summary committed (Autonomous)
44	```
45
46	## Interactive Mode (default — user present)
47
48	1. Take the user's goal / description from the skill invocation args.
49	2. Invoke `superpowers:brainstorming` with the goal as input.
50	3. Brainstorming runs its adaptive question flow — one domain-agnostic question
51	at a time. Let it drive the Q&A rhythm.
52	4. INTERCEPT POINT: when brainstorming signals readiness to chain to
53	`writing-plans`, STOP. Do NOT let it invoke `writing-plans`. Capture the
54	brainstorm output (design, requirements, decisions) into local state
55	instead. The prd-taskmaster orchestrator owns the handoff — not
56	`superpowers:brainstorming`.
57	5. Present the summary to the user for approval via `AskUserQuestion` (see
58	User Approval Gate below).
59
60	## Autonomous Mode (no user present)
61
62	Do NOT invoke `superpowers:brainstorming` — it blocks on user input and
63	will stall an unattended session. Instead, self-brainstorm using this
64	template:
65
66	1. Read the goal statement from skill args or
67	`session-context/CLAUDE-soul-purpose.md`.
68	2. Read `session-context/CLAUDE-activeContext.md` for project context.
69	3. Write discovery notes directly to
70	`session-context/discovery-{timestamp}.md` answering every question the
71	interactive flow would ask:
72	- Who is this for?
73	- What problem does it solve?
74	- What are the success metrics?
75	- What are the constraints (tech stack, timeline, team, budget,
76	integrations, regulatory)?
77	- What's explicitly out of scope?
78	- What's the scale (Solo / Team / Enterprise)?
79	4. Self-approve: the skill acts as both interrogator and approver. Document
80	assumptions explicitly so the user can audit them on wake-up.
81	5. Commit the discovery file. The git history becomes the audit trail — if
82	the user later disagrees, they can reset to that commit and re-run.
83
84	Autonomous mode is first-class, not degraded. A well-run autonomous
85	discovery produces a spec the user reads on wake-up and says "yes, that's
86	what I meant" without edits. If you find yourself needing to ask more than
87	two questions the user didn't anticipate, the discovery is under-specified —
88	stop and write a handoff note instead of proceeding.
89
90	## User Approval Gate (Interactive Mode)
91
92	After brainstorming completes, present via `AskUserQuestion`:
93
94	```
95	Discovery Complete:
96	Goal: [one sentence]
97	Audience: [who it's for]
98	Approac

People who liked this also viewed

6 picks

Type