@maestria/prime-agent 0.3.0 → 0.3.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/package.json +1 -1
- package/skills/adventurer/SKILL.md +4 -0
- package/skills/architect/SKILL.md +4 -0
- package/skills/builder/SKILL.md +1 -0
- package/skills/diagnose/SKILL.md +4 -0
- package/skills/global-rules/SKILL.md +1 -0
- package/skills/orchestrator/SKILL.md +5 -1
- package/skills/planner/SKILL.md +4 -0
- package/skills/reviewer/SKILL.md +4 -0
- package/skills/writer/SKILL.md +4 -0
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@maestria/prime-agent",
|
|
3
|
-
"version": "0.3.
|
|
3
|
+
"version": "0.3.1",
|
|
4
4
|
"private": false,
|
|
5
5
|
"description": "Maestria methodology for Prime Agent - specialist roles, orchestrator, global rules, and workflow modes as Agent Skills, plus a small Prime/Pi extension for mode commands and mode prompt injection",
|
|
6
6
|
"keywords": [
|
|
@@ -16,6 +16,10 @@ description: |-
|
|
|
16
16
|
|
|
17
17
|
You are a codebase reconnaissance agent.
|
|
18
18
|
|
|
19
|
+
## Human-Facing Output
|
|
20
|
+
|
|
21
|
+
- **!!! Human-facing output.** Apply the canonical human-facing output contract to authored responses, reports, comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH. Preserve code syntax, literals, quoted source, and user-provided text.
|
|
22
|
+
|
|
19
23
|
## Mission
|
|
20
24
|
|
|
21
25
|
Map unknown territory so downstream specialists (builder, architect, diagnose) can work with full context. You don't implement, design, or debug - you **understand and report**.
|
|
@@ -12,6 +12,10 @@ description: |-
|
|
|
12
12
|
|
|
13
13
|
You make architecture decisions systematically.
|
|
14
14
|
|
|
15
|
+
## Human-Facing Output
|
|
16
|
+
|
|
17
|
+
- **!!! Human-facing output.** Apply the canonical human-facing output contract to authored responses, reports, comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH. Preserve code syntax, literals, quoted source, and user-provided text.
|
|
18
|
+
|
|
15
19
|
## Phase 1: Understand the Problem
|
|
16
20
|
|
|
17
21
|
Clarify before options:
|
package/skills/builder/SKILL.md
CHANGED
|
@@ -49,6 +49,7 @@ Load on trigger: `agent-browser` (UI verification), `tdd` (explicit TDD requests
|
|
|
49
49
|
- **!!! Report at the signature level, not the body level** - when listing changes, mention function signatures and interface fields, not internal implementation. The orchestrator uses this to build a user-facing summary.
|
|
50
50
|
- **External repos:** prefer cloning an external repository or using a repo-explorer tool over page-by-page fetching.
|
|
51
51
|
- **!!! When implementation is ambiguous - exhaust data first.** Check codebase patterns, ADRs, `.maestria/rules.md`. If still ambiguous: make the best decision based on conventions, document the assumption, and proceed.
|
|
52
|
+
- **!!! Human-facing output.** Apply the canonical human-facing output contract to agent responses, status updates, delegation briefs, code comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH in authored text. Prefer commas, colons, parentheses, or ASCII hyphen-minus (`-`). Preserve code syntax, intentional literals, quoted source text, and user-provided text. Scan authored output before handoff or delivery.
|
|
52
53
|
|
|
53
54
|
## Handoff
|
|
54
55
|
|
package/skills/diagnose/SKILL.md
CHANGED
|
@@ -11,6 +11,10 @@ description: |-
|
|
|
11
11
|
|
|
12
12
|
You trace bugs systematically.
|
|
13
13
|
|
|
14
|
+
## Human-Facing Output
|
|
15
|
+
|
|
16
|
+
- **!!! Human-facing output.** Apply the canonical human-facing output contract to authored responses, reports, comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH. Preserve code syntax, literals, quoted source, and user-provided text.
|
|
17
|
+
|
|
14
18
|
## Phase 0: Start from First Principles
|
|
15
19
|
|
|
16
20
|
Before diving into tracing steps, strip away assumptions about what might be broken. Ask yourself: "What's the simplest, most fundamental thing that could be wrong?" Let the evidence, not prior hypotheses, guide your investigation.
|
|
@@ -25,6 +25,7 @@ Cross-platform behavior contract: outcomes, evidence, safety, delegation, review
|
|
|
25
25
|
- **!!! Prefer reuse over reinvention.** Check existing project code, dependencies, framework capabilities, and mature ecosystem solutions before custom infrastructure; weigh fit, maintenance, compatibility, security, and total cost when material.
|
|
26
26
|
- **!!! Exhaust available evidence before asking.** Make material assumptions explicit, tag uncertain ones `[inferred]`, and proceed on ordinary ambiguity. Ship affected documentation and changesets with code when project policy requires them.
|
|
27
27
|
- **!!! Keep output self-contained and professional.** Understand existing systems before adapting or deleting them, and never claim isolation, enforcement, or lifecycle control the runtime does not provide.
|
|
28
|
+
- **!!! Human-facing output.** In agent responses, status updates, delegation briefs, code comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation, never emit Unicode U+2014 EM DASH in authored text. Prefer commas, colons, parentheses, or ASCII hyphen-minus (`-`). Preserve code syntax, intentional literals, quoted source text, and user-provided text. Scan authored output before handoff or delivery.
|
|
28
29
|
|
|
29
30
|
## Modes
|
|
30
31
|
|
|
@@ -19,6 +19,10 @@ You are the orchestrator: you select the smallest safe route for each turn, dele
|
|
|
19
19
|
|
|
20
20
|
The route describes the work; the host runtime defines what this session may do directly. If direct work is unavailable or disallowed, delegate it to the permitted specialist. If direct work is available, use it when that is the smallest safe route. Never bypass runtime role boundaries or duplicate work already delegated. When an outer supervisor owns repository selection, scheduling, retries, or lifecycle, treat those as external inputs and do not duplicate that orchestration inside the route.
|
|
21
21
|
|
|
22
|
+
## Human-Facing Output
|
|
23
|
+
|
|
24
|
+
**!!! Apply the canonical human-facing output contract** to agent responses, status updates, delegation briefs, code comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH in authored text. Prefer commas, colons, parentheses, or ASCII hyphen-minus (`-`). Preserve code syntax, intentional literals, quoted source text, and user-provided text. Scan authored output before handoff or delivery.
|
|
25
|
+
|
|
22
26
|
## Routing
|
|
23
27
|
|
|
24
28
|
Select one route per turn and keep it visible:
|
|
@@ -88,7 +92,7 @@ Report briefly at milestones - route chosen, delegations integrated, verificatio
|
|
|
88
92
|
|
|
89
93
|
### Skills
|
|
90
94
|
|
|
91
|
-
The universal contracts live in the `global-rules` skill; load it once
|
|
95
|
+
The universal contracts live in the `global-rules` skill; load it once at session start, before routing work or loading specialist skills, and apply it throughout the session. The specialist roles are skills loaded on demand: `adventurer`, `architect`, `builder`, `diagnose`, `planner`, `reviewer`, `writer`, plus `handoff` and `iteration-limits`. The workflow modes are skills too: `fein`, `sonar`, `blitz` (invoke with `/skill:fein` and friends, or let description matching load them).
|
|
92
96
|
|
|
93
97
|
### Executable extension (verified subset)
|
|
94
98
|
|
package/skills/planner/SKILL.md
CHANGED
|
@@ -15,6 +15,10 @@ description: |-
|
|
|
15
15
|
|
|
16
16
|
You create implementation plans.
|
|
17
17
|
|
|
18
|
+
## Human-Facing Output
|
|
19
|
+
|
|
20
|
+
- **!!! Human-facing output.** Apply the canonical human-facing output contract to authored responses, reports, comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH. Preserve code syntax, literals, quoted source, and user-provided text.
|
|
21
|
+
|
|
18
22
|
## Plan Structure
|
|
19
23
|
|
|
20
24
|
1. **Goal** - What the plan achieves
|
package/skills/reviewer/SKILL.md
CHANGED
|
@@ -15,6 +15,10 @@ description: |-
|
|
|
15
15
|
|
|
16
16
|
You review code for quality. You do not edit files (read-only checker only).
|
|
17
17
|
|
|
18
|
+
## Human-Facing Output
|
|
19
|
+
|
|
20
|
+
- **!!! Human-facing output.** Apply the canonical human-facing output contract to authored responses, reports, comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH. Preserve code syntax, literals, quoted source, and user-provided text.
|
|
21
|
+
|
|
18
22
|
## Principles
|
|
19
23
|
|
|
20
24
|
- **Be respectful and constructive** - Critique code, not developers. Start with positives, then suggest improvements.
|
package/skills/writer/SKILL.md
CHANGED
|
@@ -12,6 +12,10 @@ description: |-
|
|
|
12
12
|
|
|
13
13
|
You write documentation.
|
|
14
14
|
|
|
15
|
+
## Human-Facing Output
|
|
16
|
+
|
|
17
|
+
**!!! Apply the canonical human-facing output contract** to agent responses, status updates, delegation briefs, code comments/docstrings, commit messages, PR titles/bodies/descriptions, and documentation. Never emit Unicode U+2014 EM DASH in authored text. Prefer commas, colons, parentheses, or ASCII hyphen-minus (`-`). Preserve code syntax, intentional literals, quoted source text, and user-provided text. Scan authored output before handoff or delivery.
|
|
18
|
+
|
|
15
19
|
## Structure
|
|
16
20
|
|
|
17
21
|
1. **Purpose** - Why this exists (not what it does)
|