@oh-my-pi/pi-coding-agent 5.4.2 → 5.5.0

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 (3) hide show
  1. package/CHANGELOG.md +5 -0
  2. package/package.json +5 -5
  3. package/src/prompts/tools/task.md +5 -1
package/CHANGELOG.md CHANGED
@@ -2,6 +2,11 @@
2
2
 
3
3
  ## [Unreleased]
4
4
 
5
+ ## [5.5.0] - 2026-01-18
6
+ ### Changed
7
+
8
+ - Updated task execution guidelines to improve prompt framing and parallelization instructions
9
+
5
10
  ## [5.4.2] - 2026-01-16
6
11
  ### Changed
7
12
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@oh-my-pi/pi-coding-agent",
3
- "version": "5.4.2",
3
+ "version": "5.5.0",
4
4
  "description": "Coding agent CLI with read, bash, edit, write tools and session management",
5
5
  "type": "module",
6
6
  "ompConfig": {
@@ -39,10 +39,10 @@
39
39
  "prepublishOnly": "bun run generate-template && bun run clean && bun run build"
40
40
  },
41
41
  "dependencies": {
42
- "@oh-my-pi/pi-agent-core": "5.4.2",
43
- "@oh-my-pi/pi-ai": "5.4.2",
44
- "@oh-my-pi/pi-git-tool": "5.4.2",
45
- "@oh-my-pi/pi-tui": "5.4.2",
42
+ "@oh-my-pi/pi-agent-core": "5.5.0",
43
+ "@oh-my-pi/pi-ai": "5.5.0",
44
+ "@oh-my-pi/pi-git-tool": "5.5.0",
45
+ "@oh-my-pi/pi-tui": "5.5.0",
46
46
  "@openai/agents": "^0.3.7",
47
47
  "@silvia-odwyer/photon-node": "^0.3.4",
48
48
  "@sinclair/typebox": "^0.34.46",
package/src/prompts/tools/task.md CHANGED
@@ -32,9 +32,13 @@ Agents with "Output: structured" have a fixed schema enforced via frontmatter; y
32
32
 
33
33
  - Always include a short description of the task in the task parameter
34
34
  - **Plan-then-execute**: Put shared constraints in `context`, keep each task focused, specify acceptance criteria; use `output` when you need structured output
35
+ - **Ask open-ended questions**: For exploration tasks, frame prompts to elicit factual discovery, not confirmation. Avoid yes/no questions that are easy to hallucinate.
36
+ - Bad: "Is there rate limiting?" or "Does the API validate tokens?" → Binary answers invite hallucination
37
+ - Good: "Find and describe how rate limiting is implemented" or "How does the API handle token validation?" → Forces investigation and factual reporting
38
+ - The subagent should report *what exists*, then YOU verify if it meets requirements
35
39
  - **Minimize tool chatter**: Avoid repeating large context; use Output tool with output ids for full logs
36
40
  - **Structured completion**: If `output` is provided, subagents must call `complete` to finish
37
- - **Parallelize**: Launch multiple agents concurrently whenever possible
41
+ - **Parallelize**: Launch multiple agents whenever possible. You MUST use a single Task call with multiple entries in the `tasks` array to do this.
38
42
  - **Isolate file scopes**: Assign each task distinct files or directories so agents don't conflict
39
43
  - **Results are intermediate data**: Agent findings provide context for YOU to perform actual work. Do not treat agent reports as "task complete" signals.
40
44
  - **Stateless invocations**: Subagents have zero memory of your conversation. Pass ALL relevant context: requirements discussed, decisions made, schemas agreed upon, file paths mentioned. If you reference something from earlier discussion without including it, the subagent will fail.