@harness-forge/cli 1.0.0 → 1.0.1
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/README.md +9 -8
- package/package.json +2 -2
package/README.md
CHANGED
|
@@ -250,7 +250,7 @@ This helps with:
|
|
|
250
250
|
- lower chance of repeating earlier mistakes
|
|
251
251
|
- cleaner handoff between humans and agents
|
|
252
252
|
|
|
253
|
-
###
|
|
253
|
+
### Will it improve decision-making?
|
|
254
254
|
|
|
255
255
|
Yes - mostly by constraining bad choices and improving context quality
|
|
256
256
|
AI agents make worse decisions when they have:
|
|
@@ -261,13 +261,6 @@ AI agents make worse decisions when they have:
|
|
|
261
261
|
- no durable task state
|
|
262
262
|
- no validation feedback loop
|
|
263
263
|
|
|
264
|
-
### For platform and enablement teams
|
|
265
|
-
|
|
266
|
-
- standardize AI-assisted workflows across repositories
|
|
267
|
-
- keep support claims explicit with a canonical capability matrix
|
|
268
|
-
- hand teams a consistent install and maintenance surface
|
|
269
|
-
- preserve auditability through generated runtime state and validation outputs
|
|
270
|
-
|
|
271
264
|
### Why it can reduce token usage
|
|
272
265
|
Harness Forge can lower token burn because it gives the agent:
|
|
273
266
|
|
|
@@ -278,6 +271,7 @@ Harness Forge can lower token burn because it gives the agent:
|
|
|
278
271
|
- repo-aware recommendations
|
|
279
272
|
- structured entrypoints instead of blind exploration
|
|
280
273
|
|
|
274
|
+
---
|
|
281
275
|
|
|
282
276
|
### For senior engineers
|
|
283
277
|
|
|
@@ -292,6 +286,13 @@ Harness Forge can lower token burn because it gives the agent:
|
|
|
292
286
|
- consume a command catalog and runtime indexes that are machine-readable
|
|
293
287
|
- route discovery via `.agents/skills/` while execution points to canonical packaged surfaces
|
|
294
288
|
|
|
289
|
+
### For platform and enablement teams
|
|
290
|
+
|
|
291
|
+
- standardize AI-assisted workflows across repositories
|
|
292
|
+
- keep support claims explicit with a canonical capability matrix
|
|
293
|
+
- hand teams a consistent install and maintenance surface
|
|
294
|
+
- preserve auditability through generated runtime state and validation outputs
|
|
295
|
+
|
|
295
296
|
---
|
|
296
297
|
|
|
297
298
|
## Advantages
|
package/package.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@harness-forge/cli",
|
|
3
|
-
"version": "1.0.
|
|
3
|
+
"version": "1.0.1",
|
|
4
4
|
"description": "Harness Forge: modular agentic AI workspace installer, catalog, and workflow runtime.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"bin": {
|
|
7
7
|
"hforge": "./dist/cli/index.js"
|
|
8
8
|
},
|
|
9
|
-
|
|
9
|
+
"repository": {
|
|
10
10
|
"type": "git",
|
|
11
11
|
"url": "git+https://github.com/ldilov/harness-forge.git"
|
|
12
12
|
},
|