agent-method 1.5.3 → 1.5.5

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.
Files changed (20) hide show
  1. package/README.md +156 -44
  2. package/lib/cli/check.js +71 -71
  3. package/lib/cli/refine.js +202 -202
  4. package/lib/cli/scan.js +28 -28
  5. package/lib/cli/status.js +61 -61
  6. package/lib/cli/upgrade.js +149 -147
  7. package/lib/init.js +296 -296
  8. package/package.json +3 -2
  9. package/templates/README.md +11 -11
  10. package/templates/entry-points/.cursorrules +11 -11
  11. package/templates/entry-points/AGENT.md +11 -11
  12. package/templates/entry-points/CLAUDE.md +11 -11
  13. package/templates/full/.cursorrules +11 -11
  14. package/templates/full/AGENT.md +11 -11
  15. package/templates/full/CLAUDE.md +11 -11
  16. package/templates/full/SESSION-LOG.md +29 -0
  17. package/templates/starter/.cursorrules +11 -11
  18. package/templates/starter/AGENT.md +11 -11
  19. package/templates/starter/CLAUDE.md +11 -11
  20. package/templates/starter/SESSION-LOG.md +29 -0
package/templates/entry-points/.cursorrules CHANGED
@@ -36,7 +36,7 @@ When a file changes, check this table and update dependent files in the same res
36
36
  | Project structure | .context/BASE.md (codebase map), this file (if new query types needed) |
37
37
  | Intelligence layer file exceeds 300 lines | Restructure into index + components subdirectory (keep active content, archive completed sections) |
38
38
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
39
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
39
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
40
40
 
41
41
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
42
42
 
@@ -66,21 +66,21 @@ tier: standard
66
66
 
67
67
  method_version: 1.5
68
68
  <!-- Tracks which methodology version generated this entry point -->
69
- <!-- Use `npx wwa status` to compare against latest -->
69
+ <!-- Use `npx agent-method status` to compare against latest -->
70
70
 
71
71
  ## CLI tools (optional)
72
72
 
73
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
73
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
74
74
 
75
75
  | When you want to... | Run |
76
76
  |---------------------|-----|
77
- | Validate this entry point | `npx wwa check` |
78
- | See what type of project this is | `npx wwa scan` |
79
- | Test how a query routes | `npx wwa route "your question"` |
80
- | Extract a refinement report | `npx wwa refine` |
81
- | Check methodology version | `npx wwa status` |
82
- | Update methodology files | `npx wwa upgrade` |
83
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
77
+ | Validate this entry point | `npx agent-method check` |
78
+ | See what type of project this is | `npx agent-method scan` |
79
+ | Test how a query routes | `npx agent-method route "your question"` |
80
+ | Extract a refinement report | `npx agent-method refine` |
81
+ | Check methodology version | `npx agent-method status` |
82
+ | Update methodology files | `npx agent-method upgrade` |
83
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
84
84
 
85
85
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
86
86
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -95,7 +95,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
95
95
  - Surface uncertainty as open questions in STATE.md — never guess silently
96
96
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
97
97
  - Propose plans and wait for approval — the human controls direction
98
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
98
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
99
99
 
100
100
  ## Do not
101
101
 
package/templates/entry-points/AGENT.md CHANGED
@@ -36,7 +36,7 @@ When a file changes, check this table and update dependent files in the same res
36
36
  | Project structure | .context/BASE.md (codebase map), this file (if new query types needed) |
37
37
  | Intelligence layer file exceeds 300 lines | Restructure into index + components subdirectory (keep active content, archive completed sections) |
38
38
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
39
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
39
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
40
40
 
41
41
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
42
42
 
@@ -66,21 +66,21 @@ tier: standard
66
66
 
67
67
  method_version: 1.5
68
68
  <!-- Tracks which methodology version generated this entry point -->
69
- <!-- Use `npx wwa status` to compare against latest -->
69
+ <!-- Use `npx agent-method status` to compare against latest -->
70
70
 
71
71
  ## CLI tools (optional)
72
72
 
73
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
73
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
74
74
 
75
75
  | When you want to... | Run |
76
76
  |---------------------|-----|
77
- | Validate this entry point | `npx wwa check` |
78
- | See what type of project this is | `npx wwa scan` |
79
- | Test how a query routes | `npx wwa route "your question"` |
80
- | Extract a refinement report | `npx wwa refine` |
81
- | Check methodology version | `npx wwa status` |
82
- | Update methodology files | `npx wwa upgrade` |
83
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
77
+ | Validate this entry point | `npx agent-method check` |
78
+ | See what type of project this is | `npx agent-method scan` |
79
+ | Test how a query routes | `npx agent-method route "your question"` |
80
+ | Extract a refinement report | `npx agent-method refine` |
81
+ | Check methodology version | `npx agent-method status` |
82
+ | Update methodology files | `npx agent-method upgrade` |
83
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
84
84
 
85
85
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
86
86
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -95,7 +95,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
95
95
  - Surface uncertainty as open questions in STATE.md — never guess silently
96
96
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
97
97
  - Propose plans and wait for approval — the human controls direction
98
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
98
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
99
99
 
100
100
  ## Do not
101
101
 
package/templates/entry-points/CLAUDE.md CHANGED
@@ -36,7 +36,7 @@ When a file changes, check this table and update dependent files in the same res
36
36
  | Project structure | .context/BASE.md (codebase map), this file (if new query types needed) |
37
37
  | Intelligence layer file exceeds 300 lines | Restructure into index + components subdirectory (keep active content, archive completed sections) |
38
38
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
39
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
39
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
40
40
 
41
41
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
42
42
 
@@ -66,21 +66,21 @@ tier: standard
66
66
 
67
67
  method_version: 1.5
68
68
  <!-- Tracks which methodology version generated this entry point -->
69
- <!-- Use `npx wwa status` to compare against latest -->
69
+ <!-- Use `npx agent-method status` to compare against latest -->
70
70
 
71
71
  ## CLI tools (optional)
72
72
 
73
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
73
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
74
74
 
75
75
  | When you want to... | Run |
76
76
  |---------------------|-----|
77
- | Validate this entry point | `npx wwa check` |
78
- | See what type of project this is | `npx wwa scan` |
79
- | Test how a query routes | `npx wwa route "your question"` |
80
- | Extract a refinement report | `npx wwa refine` |
81
- | Check methodology version | `npx wwa status` |
82
- | Update methodology files | `npx wwa upgrade` |
83
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
77
+ | Validate this entry point | `npx agent-method check` |
78
+ | See what type of project this is | `npx agent-method scan` |
79
+ | Test how a query routes | `npx agent-method route "your question"` |
80
+ | Extract a refinement report | `npx agent-method refine` |
81
+ | Check methodology version | `npx agent-method status` |
82
+ | Update methodology files | `npx agent-method upgrade` |
83
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
84
84
 
85
85
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
86
86
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -95,7 +95,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
95
95
  - Surface uncertainty as open questions in STATE.md — never guess silently
96
96
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
97
97
  - Propose plans and wait for approval — the human controls direction
98
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
98
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
99
99
 
100
100
  ## Do not
101
101
 
package/templates/full/.cursorrules CHANGED
@@ -41,7 +41,7 @@ When a file changes, check this table and update dependent files in the same res
41
41
  | File split, created, deleted, or renamed | .context/REGISTRY.md (file tree, topic index, structural log) |
42
42
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
43
43
  | Lifecycle stage change | PROJECT-PROFILE.md (update stage + date), STATE.md (record decision) |
44
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
44
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
45
45
 
46
46
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
47
47
 
@@ -83,21 +83,21 @@ tier: full
83
83
 
84
84
  method_version: 1.5
85
85
  <!-- Tracks which methodology version generated this entry point -->
86
- <!-- Use `npx wwa status` to compare against latest -->
86
+ <!-- Use `npx agent-method status` to compare against latest -->
87
87
 
88
88
  ## CLI tools (optional)
89
89
 
90
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
90
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
91
91
 
92
92
  | When you want to... | Run |
93
93
  |---------------------|-----|
94
- | Validate this entry point | `npx wwa check` |
95
- | See what type of project this is | `npx wwa scan` |
96
- | Test how a query routes | `npx wwa route "your question"` |
97
- | Extract a refinement report | `npx wwa refine` |
98
- | Check methodology version | `npx wwa status` |
99
- | Update methodology files | `npx wwa upgrade` |
100
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
94
+ | Validate this entry point | `npx agent-method check` |
95
+ | See what type of project this is | `npx agent-method scan` |
96
+ | Test how a query routes | `npx agent-method route "your question"` |
97
+ | Extract a refinement report | `npx agent-method refine` |
98
+ | Check methodology version | `npx agent-method status` |
99
+ | Update methodology files | `npx agent-method upgrade` |
100
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
101
101
 
102
102
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
103
103
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -113,7 +113,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
113
113
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
114
114
  - Propose plans and wait for approval — the human controls direction
115
115
  - SUMMARY.md entries follow audit trail format: date, plan, outcome, files, decisions, next
116
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
116
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
117
117
 
118
118
  ## Do not
119
119
 
package/templates/full/AGENT.md CHANGED
@@ -41,7 +41,7 @@ When a file changes, check this table and update dependent files in the same res
41
41
  | File split, created, deleted, or renamed | .context/REGISTRY.md (file tree, topic index, structural log) |
42
42
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
43
43
  | Lifecycle stage change | PROJECT-PROFILE.md (update stage + date), STATE.md (record decision) |
44
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
44
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
45
45
 
46
46
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
47
47
 
@@ -83,21 +83,21 @@ tier: full
83
83
 
84
84
  method_version: 1.5
85
85
  <!-- Tracks which methodology version generated this entry point -->
86
- <!-- Use `npx wwa status` to compare against latest -->
86
+ <!-- Use `npx agent-method status` to compare against latest -->
87
87
 
88
88
  ## CLI tools (optional)
89
89
 
90
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
90
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
91
91
 
92
92
  | When you want to... | Run |
93
93
  |---------------------|-----|
94
- | Validate this entry point | `npx wwa check` |
95
- | See what type of project this is | `npx wwa scan` |
96
- | Test how a query routes | `npx wwa route "your question"` |
97
- | Extract a refinement report | `npx wwa refine` |
98
- | Check methodology version | `npx wwa status` |
99
- | Update methodology files | `npx wwa upgrade` |
100
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
94
+ | Validate this entry point | `npx agent-method check` |
95
+ | See what type of project this is | `npx agent-method scan` |
96
+ | Test how a query routes | `npx agent-method route "your question"` |
97
+ | Extract a refinement report | `npx agent-method refine` |
98
+ | Check methodology version | `npx agent-method status` |
99
+ | Update methodology files | `npx agent-method upgrade` |
100
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
101
101
 
102
102
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
103
103
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -113,7 +113,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
113
113
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
114
114
  - Propose plans and wait for approval — the human controls direction
115
115
  - SUMMARY.md entries follow audit trail format: date, plan, outcome, files, decisions, next
116
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
116
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
117
117
 
118
118
  ## Do not
119
119
 
package/templates/full/CLAUDE.md CHANGED
@@ -41,7 +41,7 @@ When a file changes, check this table and update dependent files in the same res
41
41
  | File split, created, deleted, or renamed | .context/REGISTRY.md (file tree, topic index, structural log) |
42
42
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
43
43
  | Lifecycle stage change | PROJECT-PROFILE.md (update stage + date), STATE.md (record decision) |
44
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
44
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
45
45
 
46
46
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
47
47
 
@@ -83,21 +83,21 @@ tier: full
83
83
 
84
84
  method_version: 1.5
85
85
  <!-- Tracks which methodology version generated this entry point -->
86
- <!-- Use `npx wwa status` to compare against latest -->
86
+ <!-- Use `npx agent-method status` to compare against latest -->
87
87
 
88
88
  ## CLI tools (optional)
89
89
 
90
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
90
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
91
91
 
92
92
  | When you want to... | Run |
93
93
  |---------------------|-----|
94
- | Validate this entry point | `npx wwa check` |
95
- | See what type of project this is | `npx wwa scan` |
96
- | Test how a query routes | `npx wwa route "your question"` |
97
- | Extract a refinement report | `npx wwa refine` |
98
- | Check methodology version | `npx wwa status` |
99
- | Update methodology files | `npx wwa upgrade` |
100
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
94
+ | Validate this entry point | `npx agent-method check` |
95
+ | See what type of project this is | `npx agent-method scan` |
96
+ | Test how a query routes | `npx agent-method route "your question"` |
97
+ | Extract a refinement report | `npx agent-method refine` |
98
+ | Check methodology version | `npx agent-method status` |
99
+ | Update methodology files | `npx agent-method upgrade` |
100
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
101
101
 
102
102
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
103
103
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -113,7 +113,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
113
113
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
114
114
  - Propose plans and wait for approval — the human controls direction
115
115
  - SUMMARY.md entries follow audit trail format: date, plan, outcome, files, decisions, next
116
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
116
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
117
117
 
118
118
  ## Do not
119
119
 
package/templates/full/SESSION-LOG.md CHANGED
@@ -40,6 +40,27 @@ Append-only session observation log for case study data collection. Each session
40
40
  - **high**: Entry point + STATE.md + specialist context + multiple project files
41
41
  - **very high**: Extensive project context, multiple specialists, cross-file analysis
42
42
 
43
+ **User response** (agent-observed — how did the user respond to the result?):
44
+ - **accepted**: User proceeded to next step without changes
45
+ - **edited**: User manually modified the agent's output
46
+ - **revised**: User asked agent to redo or revise the result
47
+ - **rejected**: User said no or declined the result entirely
48
+ - **redirected**: User changed approach or gave new instructions
49
+
50
+ **Refinement magnitude** (for medium/high effort only — how much changed from first attempt to final?):
51
+ - **none**: Accepted as-is, 0% changed
52
+ - **minor**: Small fixes — typos, naming, formatting (<10% changed)
53
+ - **moderate**: Logic or structural changes, added/removed sections (10–50% changed)
54
+ - **major**: Significant rework of approach or content (50–80% changed)
55
+ - **rework**: Mostly rewritten, original approach abandoned (>80% changed)
56
+
57
+ **Delta categories** (what kinds of changes were needed — select all that apply):
58
+ - **accuracy**: Factual errors or incorrect implementation
59
+ - **completeness**: Missing parts, incomplete coverage
60
+ - **approach**: Wrong method or strategy
61
+ - **scope**: Over-scoped or under-scoped
62
+ - **style**: Formatting, naming, conventions
63
+
43
64
  ## Observation checklist
44
65
 
45
66
  At session close (or high-effort task completion), reflect on these before writing the entry:
@@ -54,6 +75,10 @@ At session close (or high-effort task completion), reflect on these before writi
54
75
  9. How ambiguous was the user's request? (low / medium / high)
55
76
  10. How much project context was loaded? (very low / low / medium / high / very high)
56
77
  11. Approximate token usage and time spent?
78
+ 12. How did the user respond to the result? (accepted / edited / revised / rejected / redirected)
79
+ 13. For medium/high effort: how many revision cycles before acceptance?
80
+ 14. What magnitude of change between first attempt and final result? (none / minor / moderate / major / rework)
81
+ 15. What categories of refinement were needed? (accuracy / completeness / approach / scope / style)
57
82
 
58
83
  ## Entry format
59
84
 
@@ -66,6 +91,10 @@ Tokens: ~{N}k | Time: ~{N} min
66
91
  Queries: {query types encountered}
67
92
  Features: {feature IDs activated}
68
93
  Cascades: {triggered}/{expected} | Decisions: {count}
94
+ Response: {accepted / edited / revised / rejected / redirected}
95
+ Revisions: {0 | count of revision cycles} | Magnitude: {none / minor / moderate / major / rework}
96
+ Delta: {n/a | categories: accuracy, completeness, approach, scope, style} | Survival: ~{N}%
97
+ Delta notes: {n/a | brief description of what changed between first attempt and final}
69
98
  Friction: {none | brief description}
70
99
  Finding: {none | observation with methodology implication}
71
100
  -->
package/templates/starter/.cursorrules CHANGED
@@ -39,7 +39,7 @@ When a file changes, check this table and update dependent files in the same res
39
39
  | Intelligence layer file exceeds 300 lines | Restructure into index + components subdirectory (keep active content, archive completed sections) |
40
40
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
41
41
  | Lifecycle stage change | PROJECT-PROFILE.md (update stage + date), STATE.md (record decision) |
42
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
42
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
43
43
 
44
44
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
45
45
 
@@ -70,21 +70,21 @@ tier: standard
70
70
 
71
71
  method_version: 1.5
72
72
  <!-- Tracks which methodology version generated this entry point -->
73
- <!-- Use `npx wwa status` to compare against latest -->
73
+ <!-- Use `npx agent-method status` to compare against latest -->
74
74
 
75
75
  ## CLI tools (optional)
76
76
 
77
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
77
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
78
78
 
79
79
  | When you want to... | Run |
80
80
  |---------------------|-----|
81
- | Validate this entry point | `npx wwa check` |
82
- | See what type of project this is | `npx wwa scan` |
83
- | Test how a query routes | `npx wwa route "your question"` |
84
- | Extract a refinement report | `npx wwa refine` |
85
- | Check methodology version | `npx wwa status` |
86
- | Update methodology files | `npx wwa upgrade` |
87
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
81
+ | Validate this entry point | `npx agent-method check` |
82
+ | See what type of project this is | `npx agent-method scan` |
83
+ | Test how a query routes | `npx agent-method route "your question"` |
84
+ | Extract a refinement report | `npx agent-method refine` |
85
+ | Check methodology version | `npx agent-method status` |
86
+ | Update methodology files | `npx agent-method upgrade` |
87
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
88
88
 
89
89
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
90
90
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -99,7 +99,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
99
99
  - Surface uncertainty as open questions in STATE.md — never guess silently
100
100
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
101
101
  - Propose plans and wait for approval — the human controls direction
102
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
102
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
103
103
 
104
104
  ## Do not
105
105
 
package/templates/starter/AGENT.md CHANGED
@@ -39,7 +39,7 @@ When a file changes, check this table and update dependent files in the same res
39
39
  | Intelligence layer file exceeds 300 lines | Restructure into index + components subdirectory (keep active content, archive completed sections) |
40
40
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
41
41
  | Lifecycle stage change | PROJECT-PROFILE.md (update stage + date), STATE.md (record decision) |
42
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
42
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
43
43
 
44
44
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
45
45
 
@@ -70,21 +70,21 @@ tier: standard
70
70
 
71
71
  method_version: 1.5
72
72
  <!-- Tracks which methodology version generated this entry point -->
73
- <!-- Use `npx wwa status` to compare against latest -->
73
+ <!-- Use `npx agent-method status` to compare against latest -->
74
74
 
75
75
  ## CLI tools (optional)
76
76
 
77
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
77
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
78
78
 
79
79
  | When you want to... | Run |
80
80
  |---------------------|-----|
81
- | Validate this entry point | `npx wwa check` |
82
- | See what type of project this is | `npx wwa scan` |
83
- | Test how a query routes | `npx wwa route "your question"` |
84
- | Extract a refinement report | `npx wwa refine` |
85
- | Check methodology version | `npx wwa status` |
86
- | Update methodology files | `npx wwa upgrade` |
87
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
81
+ | Validate this entry point | `npx agent-method check` |
82
+ | See what type of project this is | `npx agent-method scan` |
83
+ | Test how a query routes | `npx agent-method route "your question"` |
84
+ | Extract a refinement report | `npx agent-method refine` |
85
+ | Check methodology version | `npx agent-method status` |
86
+ | Update methodology files | `npx agent-method upgrade` |
87
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
88
88
 
89
89
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
90
90
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -99,7 +99,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
99
99
  - Surface uncertainty as open questions in STATE.md — never guess silently
100
100
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
101
101
  - Propose plans and wait for approval — the human controls direction
102
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
102
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
103
103
 
104
104
  ## Do not
105
105
 
package/templates/starter/CLAUDE.md CHANGED
@@ -39,7 +39,7 @@ When a file changes, check this table and update dependent files in the same res
39
39
  | Intelligence layer file exceeds 300 lines | Restructure into index + components subdirectory (keep active content, archive completed sections) |
40
40
  | New domain area | .context/BASE.md, consider new .context/ specialist, this file (if new scoping row) |
41
41
  | Lifecycle stage change | PROJECT-PROFILE.md (update stage + date), STATE.md (record decision) |
42
- | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, workflow, features, cascades, friction, findings) |
42
+ | Session close or high-effort task completion | SESSION-LOG.md (append metrics entry — effort, ambiguity, context level, tokens, time, user response, refinement delta, workflow, features, cascades, friction, findings) |
43
43
 
44
44
  <!-- INSTRUCTION: Add project-specific cascade rules below the universal ones above. -->
45
45
 
@@ -70,21 +70,21 @@ tier: standard
70
70
 
71
71
  method_version: 1.5
72
72
  <!-- Tracks which methodology version generated this entry point -->
73
- <!-- Use `npx wwa status` to compare against latest -->
73
+ <!-- Use `npx agent-method status` to compare against latest -->
74
74
 
75
75
  ## CLI tools (optional)
76
76
 
77
- Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
77
+ Available via `npx agent-method` (zero-install) or `pip install wwa-tools`:
78
78
 
79
79
  | When you want to... | Run |
80
80
  |---------------------|-----|
81
- | Validate this entry point | `npx wwa check` |
82
- | See what type of project this is | `npx wwa scan` |
83
- | Test how a query routes | `npx wwa route "your question"` |
84
- | Extract a refinement report | `npx wwa refine` |
85
- | Check methodology version | `npx wwa status` |
86
- | Update methodology files | `npx wwa upgrade` |
87
- | See what an entry point should contain | `npx wwa init code` / `context` / `data` / `mix` |
81
+ | Validate this entry point | `npx agent-method check` |
82
+ | See what type of project this is | `npx agent-method scan` |
83
+ | Test how a query routes | `npx agent-method route "your question"` |
84
+ | Extract a refinement report | `npx agent-method refine` |
85
+ | Check methodology version | `npx agent-method status` |
86
+ | Update methodology files | `npx agent-method upgrade` |
87
+ | See what an entry point should contain | `npx agent-method init code` / `context` / `data` / `mix` |
88
88
 
89
89
  <!-- INSTRUCTION: The agent can suggest these commands when the user asks about validation,
90
90
  project setup, or methodology updates. All commands auto-detect project type and find
@@ -99,7 +99,7 @@ Available via `npx wwa` (zero-install) or `pip install wwa-tools`:
99
99
  - Surface uncertainty as open questions in STATE.md — never guess silently
100
100
  - Keep intelligence layer files under 300 lines — split into index + components subdirectory when exceeded
101
101
  - Propose plans and wait for approval — the human controls direction
102
- - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, and time. Never skip, never read previous entries during normal work
102
+ - At session close or after any high-effort task, append a metrics entry to SESSION-LOG.md — include effort level, question ambiguity, context level, estimated tokens, time, user response (accepted/edited/revised/rejected/redirected), and for medium/high effort tasks: revision count, refinement magnitude (none/minor/moderate/major/rework), delta categories, and survival rate. Never skip, never read previous entries during normal work
103
103
 
104
104
  ## Do not
105
105
 
package/templates/starter/SESSION-LOG.md CHANGED
@@ -40,6 +40,27 @@ Append-only session observation log for case study data collection. Each session
40
40
  - **high**: Entry point + STATE.md + specialist context + multiple project files
41
41
  - **very high**: Extensive project context, multiple specialists, cross-file analysis
42
42
 
43
+ **User response** (agent-observed — how did the user respond to the result?):
44
+ - **accepted**: User proceeded to next step without changes
45
+ - **edited**: User manually modified the agent's output
46
+ - **revised**: User asked agent to redo or revise the result
47
+ - **rejected**: User said no or declined the result entirely
48
+ - **redirected**: User changed approach or gave new instructions
49
+
50
+ **Refinement magnitude** (for medium/high effort only — how much changed from first attempt to final?):
51
+ - **none**: Accepted as-is, 0% changed
52
+ - **minor**: Small fixes — typos, naming, formatting (<10% changed)
53
+ - **moderate**: Logic or structural changes, added/removed sections (10–50% changed)
54
+ - **major**: Significant rework of approach or content (50–80% changed)
55
+ - **rework**: Mostly rewritten, original approach abandoned (>80% changed)
56
+
57
+ **Delta categories** (what kinds of changes were needed — select all that apply):
58
+ - **accuracy**: Factual errors or incorrect implementation
59
+ - **completeness**: Missing parts, incomplete coverage
60
+ - **approach**: Wrong method or strategy
61
+ - **scope**: Over-scoped or under-scoped
62
+ - **style**: Formatting, naming, conventions
63
+
43
64
  ## Observation checklist
44
65
 
45
66
  At session close (or high-effort task completion), reflect on these before writing the entry:
@@ -54,6 +75,10 @@ At session close (or high-effort task completion), reflect on these before writi
54
75
  9. How ambiguous was the user's request? (low / medium / high)
55
76
  10. How much project context was loaded? (very low / low / medium / high / very high)
56
77
  11. Approximate token usage and time spent?
78
+ 12. How did the user respond to the result? (accepted / edited / revised / rejected / redirected)
79
+ 13. For medium/high effort: how many revision cycles before acceptance?
80
+ 14. What magnitude of change between first attempt and final result? (none / minor / moderate / major / rework)
81
+ 15. What categories of refinement were needed? (accuracy / completeness / approach / scope / style)
57
82
 
58
83
  ## Entry format
59
84
 
@@ -66,6 +91,10 @@ Tokens: ~{N}k | Time: ~{N} min
66
91
  Queries: {query types encountered}
67
92
  Features: {feature IDs activated}
68
93
  Cascades: {triggered}/{expected} | Decisions: {count}
94
+ Response: {accepted / edited / revised / rejected / redirected}
95
+ Revisions: {0 | count of revision cycles} | Magnitude: {none / minor / moderate / major / rework}
96
+ Delta: {n/a | categories: accuracy, completeness, approach, scope, style} | Survival: ~{N}%
97
+ Delta notes: {n/a | brief description of what changed between first attempt and final}
69
98
  Friction: {none | brief description}
70
99
  Finding: {none | observation with methodology implication}
71
100
  -->