@chrono-meta/fh-gate 1.4.27 → 1.4.28
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
CHANGED
|
@@ -85,7 +85,7 @@ Execution method · agent selection · recording location → infer and proceed
|
|
|
85
85
|
| "Do X" — target and completion criteria are clear | Clear | Immediately enter Step 0-b |
|
|
86
86
|
| "Improve X" — improvement dimension unclear | Direction unclear | Ask for clarification |
|
|
87
87
|
| "Something feels off" — diagnosis target unclear | Execution unclear → infer | Auto-dispatch harness-doctor |
|
|
88
|
-
| "
|
|
88
|
+
| "Wrap up / harvest this session" — trigger is clear | Clear | Run harvest-loop based on recent work ("run the pipeline" itself is ceded to pipeline-conductor — route end-to-end sweeps there, not here) |
|
|
89
89
|
| No request — session context exists | Execution unclear → infer | Auto-compose based on CATALOG/MEMORY, then confirm |
|
|
90
90
|
|
|
91
91
|
> **Detail**: See `SKILL_detail.md §Clarification` — 2-question format, 4-question structured confirmation, meta-prompting intervention — read when direction ambiguity requires a clarification block.
|
|
@@ -13,117 +13,18 @@ deprecated_date: 2026-06-02
|
|
|
13
13
|
successor: harness-doctor
|
|
14
14
|
---
|
|
15
15
|
|
|
16
|
-
# self-marketing-lint —
|
|
16
|
+
# self-marketing-lint — DEPRECATED
|
|
17
17
|
|
|
18
|
-
|
|
19
|
-
|
|
18
|
+
This skill was absorbed into **harness-doctor** on 2026-06-02. Its self-marketing /
|
|
19
|
+
cushion-word / version-brag detection patterns now live in **harness-doctor Step 3-L
|
|
20
|
+
(`--lint` mode)**, which carries the canonical baseline.
|
|
20
21
|
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
- `/self-marketing-lint`
|
|
24
|
-
- "Check FH files", "remove marketing language", "description diet"
|
|
25
|
-
- "Catch self-marketing language", "run the lint"
|
|
26
|
-
- When surfaced as a HIGH item in harvest-loop (P10 series)
|
|
27
|
-
|
|
28
|
-
## Detection Patterns (Inline Baseline)
|
|
29
|
-
|
|
30
|
-
The following criteria are self-contained inside this skill. No external personal-memory file is required.
|
|
31
|
-
|
|
32
|
-
### Self-Marketing Anti-Patterns
|
|
33
|
-
|
|
34
|
-
1. **Self-declarations of quality**: "industry-leading", "world-class", "production-ready", "the world's first" without external evidence
|
|
35
|
-
2. **Round-counters and version brags**: "after 26 iterations", "v0.5 maturity", "22nd naming round", "prototype 9th iteration" — irrelevant to function
|
|
36
|
-
3. **Cushion language**: "perfectly", "easily", "seamlessly", "instantly", "without burden" — content-free intensifiers
|
|
37
|
-
4. **Owner self-reference loops**: "validated by our team", "proven internally" — circular evidence with no external check
|
|
38
|
-
5. **Marketing word stacking**: 3+ adjectives in one sentence
|
|
39
|
-
6. **Spec-only artifacts**: skills/agents missing a Done When section — declaration without observable output
|
|
40
|
-
|
|
41
|
-
| Pattern type | Examples (detection targets) | Replacement direction |
|
|
42
|
-
|---|---|---|
|
|
43
|
-
| Version labels | "v0.3", "latest version", "(added 2026-05-24)" | Remove (git history manages versioning) |
|
|
44
|
-
| Emphasis words | "fully automated", "instantly", "automatically", "full coverage" | Replace with functional description |
|
|
45
|
-
| Iteration counts | "22nd naming", "26th round", "prototype 9th iteration" | Remove or move to external document |
|
|
46
|
-
| Self-declarations | "This skill is the world's first", "Core FH skill" | Delete |
|
|
47
|
-
| Cushion language | "easily", "conveniently", "without burden" | Replace with behavior-based description |
|
|
48
|
-
| Spec-only (no outputs) | Skills missing Done When | Require adding an outputs section |
|
|
49
|
-
|
|
50
|
-
## Execution Steps
|
|
51
|
-
|
|
52
|
-
### Step 1. Scan Target Files
|
|
53
|
-
|
|
54
|
-
```bash
|
|
55
|
-
# full list of skill files
|
|
56
|
-
find {FH_ROOT}/plugins -name "SKILL.md" | sort
|
|
57
|
-
|
|
58
|
-
# full list of agent files
|
|
59
|
-
find {FH_ROOT}/.claude/agents -name "*.md" | sort
|
|
60
|
-
|
|
61
|
-
# docs/ public documents
|
|
62
|
-
find {FH_ROOT}/docs -name "*.md" | sort
|
|
63
|
-
```
|
|
64
|
-
|
|
65
|
-
Scope: specified files only if target is given / full scan if unspecified.
|
|
66
|
-
|
|
67
|
-
### Step 2. Confirm Inline Baseline
|
|
68
|
-
|
|
69
|
-
Re-read the "Detection Patterns (Inline Baseline)" section above. No external file load is required — the skill is self-contained.
|
|
70
|
-
|
|
71
|
-
### Step 3. Pattern Detection
|
|
72
|
-
|
|
73
|
-
Grep detection patterns in each file's description field + body.
|
|
74
|
-
|
|
75
|
-
```bash
|
|
76
|
-
# version / iteration count patterns
|
|
77
|
-
grep -n "v[0-9]\+\.[0-9]\|[0-9]\+th\|[0-9]\+th iteration\|prototype [0-9]\+" {file}
|
|
78
|
-
|
|
79
|
-
# emphasis word patterns
|
|
80
|
-
grep -n "fully\|instantly\|full coverage\|automatically\|conveniently\|easily\|without burden" {file}
|
|
81
|
-
|
|
82
|
-
# self-declaration patterns
|
|
83
|
-
grep -n "core skill\|world's first\|best\|most\|top" {file}
|
|
84
|
-
```
|
|
85
|
-
|
|
86
|
-
### Step 4. Output Replacement Suggestions
|
|
87
|
-
|
|
88
|
-
Organize and output detection results per file.
|
|
89
|
-
|
|
90
|
-
```
|
|
91
|
-
File: {path}
|
|
92
|
-
L{line}: "{original text}"
|
|
93
|
-
→ Suggested: "{replacement text}" (reason: {pattern type})
|
|
94
|
-
→ Or: recommend removal (reason: {pattern type})
|
|
95
|
-
```
|
|
96
|
-
|
|
97
|
-
**No automatic edits**: Output suggestions only. Actual edits proceed after user confirmation.
|
|
98
|
-
|
|
99
|
-
### Step 5. Summary Report
|
|
22
|
+
**Use instead:**
|
|
100
23
|
|
|
101
24
|
```
|
|
102
|
-
|
|
103
|
-
|
|
104
|
-
Target: {N} files
|
|
105
|
-
Detected: {N} items
|
|
106
|
-
|
|
107
|
-
| File | Detection count | Most common pattern |
|
|
108
|
-
|---|---|---|
|
|
109
|
-
...
|
|
110
|
-
|
|
111
|
-
Recommendation: prioritize files with highest detection counts for review
|
|
25
|
+
/harness-doctor --lint
|
|
112
26
|
```
|
|
113
27
|
|
|
114
|
-
|
|
115
|
-
|
|
116
|
-
|
|
117
|
-
1. Full scan of target files complete
|
|
118
|
-
2. Inline baseline criteria applied (no external file dependency)
|
|
119
|
-
3. Detection results organized and output per file
|
|
120
|
-
4. No automatic edits — user confirmation gate maintained
|
|
121
|
-
```
|
|
122
|
-
|
|
123
|
-
## Connections
|
|
124
|
-
|
|
125
|
-
| Situation | Connection |
|
|
126
|
-
|---|---|
|
|
127
|
-
| harvest-loop HIGH P10 series triggered | auto-suggest `/self-marketing-lint` |
|
|
128
|
-
| During cold audit | `/steel-quench` + self-marketing-lint in parallel |
|
|
129
|
-
| Before external distribution | `fh-meta:hub-persona-auditor` followed by self-marketing-lint |
|
|
28
|
+
Triggers that previously reached this skill (`check FH files`, `remove marketing
|
|
29
|
+
language`, `description diet`) now route to harness-doctor Step 3-L. This stub remains
|
|
30
|
+
only so old references resolve.
|