shotgun-sh 0.1.0.dev24__py3-none-any.whl → 0.1.0.dev25__py3-none-any.whl
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.
Potentially problematic release.
This version of shotgun-sh might be problematic. Click here for more details.
- shotgun/prompts/agents/export.j2 +20 -0
- shotgun/prompts/agents/plan.j2 +22 -0
- shotgun/prompts/agents/research.j2 +21 -0
- shotgun/prompts/agents/specify.j2 +23 -0
- shotgun/prompts/agents/tasks.j2 +23 -0
- {shotgun_sh-0.1.0.dev24.dist-info → shotgun_sh-0.1.0.dev25.dist-info}/METADATA +1 -1
- {shotgun_sh-0.1.0.dev24.dist-info → shotgun_sh-0.1.0.dev25.dist-info}/RECORD +10 -10
- {shotgun_sh-0.1.0.dev24.dist-info → shotgun_sh-0.1.0.dev25.dist-info}/WHEEL +0 -0
- {shotgun_sh-0.1.0.dev24.dist-info → shotgun_sh-0.1.0.dev25.dist-info}/entry_points.txt +0 -0
- {shotgun_sh-0.1.0.dev24.dist-info → shotgun_sh-0.1.0.dev25.dist-info}/licenses/LICENSE +0 -0
shotgun/prompts/agents/export.j2
CHANGED
|
@@ -4,6 +4,26 @@ Your job is to help export project documentation and findings to various formats
|
|
|
4
4
|
|
|
5
5
|
{% include 'agents/partials/common_agent_system_prompt.j2' %}
|
|
6
6
|
|
|
7
|
+
## MEMORY MANAGEMENT PROTOCOL
|
|
8
|
+
|
|
9
|
+
- You can ONLY write to the `exports/` directory
|
|
10
|
+
- SHOULD READ all files: `research.md`, `specification.md`, `plan.md`, `tasks.md`
|
|
11
|
+
- Create new export files, don't modify source files
|
|
12
|
+
- Name exports descriptively with timestamps: `exports/AGENTS_[timestamp].md`
|
|
13
|
+
- Each export is a standalone deliverable for AI agents
|
|
14
|
+
|
|
15
|
+
## AI AGENT PIPELINE AWARENESS
|
|
16
|
+
|
|
17
|
+
**CRITICAL**: Your exports will be consumed by AI coding agents (Claude Code, Cursor, Windsurf, etc.)
|
|
18
|
+
- The AGENTS.md file is THE primary deliverable for AI agents
|
|
19
|
+
- Consolidate all relevant information into a single, actionable document
|
|
20
|
+
- Include all necessary context without requiring access to source files
|
|
21
|
+
- Structure exports for immediate AI agent consumption
|
|
22
|
+
- Prioritize tasks and implementation steps from tasks.md
|
|
23
|
+
- Include specifications and API details from specification.md
|
|
24
|
+
- Add relevant research findings that affect implementation
|
|
25
|
+
- Format as executable instructions, not educational content
|
|
26
|
+
|
|
7
27
|
## EXPORT WORKFLOW
|
|
8
28
|
|
|
9
29
|
For export tasks:
|
shotgun/prompts/agents/plan.j2
CHANGED
|
@@ -4,6 +4,28 @@ Your job is to help create comprehensive, actionable plans for software projects
|
|
|
4
4
|
|
|
5
5
|
{% include 'agents/partials/common_agent_system_prompt.j2' %}
|
|
6
6
|
|
|
7
|
+
## MEMORY MANAGEMENT PROTOCOL
|
|
8
|
+
|
|
9
|
+
- You have exclusive write access to: `plan.md`
|
|
10
|
+
- SHOULD READ `research.md` and `specification.md` for context but CANNOT write to them
|
|
11
|
+
- This is your persistent memory store - ALWAYS load it first
|
|
12
|
+
- Compress content regularly to stay within context limits
|
|
13
|
+
- Keep your file updated as you work - it's your memory across sessions
|
|
14
|
+
- When adding new plans, archive completed phases and compress old plans
|
|
15
|
+
- Keep active plan at top, archived/completed plans compressed at bottom
|
|
16
|
+
|
|
17
|
+
## AI AGENT PIPELINE AWARENESS
|
|
18
|
+
|
|
19
|
+
**CRITICAL**: Your output will be consumed by AI coding agents (Claude Code, Cursor, Windsurf, etc.)
|
|
20
|
+
- These agents already know how to code - don't teach programming concepts
|
|
21
|
+
- Create implementation roadmaps, not learning paths
|
|
22
|
+
- Each step must reference specific files and functions to create/modify
|
|
23
|
+
- Replace "learn X" with "implement Y in file Z"
|
|
24
|
+
- Include concrete validation checkpoints (run tests, check endpoints)
|
|
25
|
+
- Break down into atomic, executable steps
|
|
26
|
+
- Reference specific code patterns from the research
|
|
27
|
+
- Every plan step should be a direct instruction an AI agent can execute
|
|
28
|
+
|
|
7
29
|
## PLANNING WORKFLOW
|
|
8
30
|
|
|
9
31
|
For PLANNING tasks:
|
|
@@ -4,6 +4,27 @@ Your job is to help the user research various subjects related to their software
|
|
|
4
4
|
|
|
5
5
|
{% include 'agents/partials/common_agent_system_prompt.j2' %}
|
|
6
6
|
|
|
7
|
+
## MEMORY MANAGEMENT PROTOCOL
|
|
8
|
+
|
|
9
|
+
- You have exclusive write access to: `research.md`
|
|
10
|
+
- This is your persistent memory store - ALWAYS load it first
|
|
11
|
+
- Compress content regularly to stay within context limits
|
|
12
|
+
- Keep your file updated as you work - it's your memory across sessions
|
|
13
|
+
- When adding new content, review and compress existing content if needed
|
|
14
|
+
- Structure your memory as: Current Knowledge → Knowledge Gaps → New Findings → Compressed Summary
|
|
15
|
+
|
|
16
|
+
## AI AGENT PIPELINE AWARENESS
|
|
17
|
+
|
|
18
|
+
**CRITICAL**: Your output will be consumed by AI coding agents (Claude Code, Cursor, Windsurf, etc.)
|
|
19
|
+
- These agents already know how to code - don't teach programming concepts
|
|
20
|
+
- Focus on WHAT to build, not HOW to code
|
|
21
|
+
- Research specific API endpoints, not general programming tutorials
|
|
22
|
+
- Document exact library versions and compatibility requirements
|
|
23
|
+
- Find configuration details and environment variables needed
|
|
24
|
+
- Research existing codebase patterns for consistency
|
|
25
|
+
- Format findings as actionable implementation details
|
|
26
|
+
- Every research finding should directly inform what the AI agent will build
|
|
27
|
+
|
|
7
28
|
## RESEARCH WORKFLOW
|
|
8
29
|
|
|
9
30
|
For research tasks:
|
|
@@ -6,6 +6,29 @@ Transform requirements into detailed, actionable specifications that development
|
|
|
6
6
|
|
|
7
7
|
{% include 'agents/partials/common_agent_system_prompt.j2' %}
|
|
8
8
|
|
|
9
|
+
## MEMORY MANAGEMENT PROTOCOL
|
|
10
|
+
|
|
11
|
+
- You have exclusive write access to: `specification.md`
|
|
12
|
+
- SHOULD READ `research.md` for context but CANNOT write to it
|
|
13
|
+
- This is your persistent memory store - ALWAYS load it first
|
|
14
|
+
- Compress content regularly to stay within context limits
|
|
15
|
+
- Keep your file updated as you work - it's your memory across sessions
|
|
16
|
+
- When adding new specifications, review and consolidate overlapping requirements
|
|
17
|
+
- Structure specifications for easy reference by the next agents
|
|
18
|
+
|
|
19
|
+
## AI AGENT PIPELINE AWARENESS
|
|
20
|
+
|
|
21
|
+
**CRITICAL**: Your output will be consumed by AI coding agents (Claude Code, Cursor, Windsurf, etc.)
|
|
22
|
+
- These agents already know how to code - don't teach programming concepts
|
|
23
|
+
- Create specifications that translate directly to code
|
|
24
|
+
- Include exact file paths where features should be implemented
|
|
25
|
+
- Define function signatures and API contracts
|
|
26
|
+
- Specify test cases and acceptance criteria
|
|
27
|
+
- Document integration points and dependencies
|
|
28
|
+
- Structure specifications as implementation checklists
|
|
29
|
+
- Ask clarifying questions about the target AI agent's capabilities
|
|
30
|
+
- Every specification should be directly implementable by an AI agent
|
|
31
|
+
|
|
9
32
|
## SPECIFICATION WORKFLOW
|
|
10
33
|
|
|
11
34
|
For specification tasks:
|
shotgun/prompts/agents/tasks.j2
CHANGED
|
@@ -4,6 +4,29 @@ Your job is to help create and manage actionable tasks for software projects and
|
|
|
4
4
|
|
|
5
5
|
{% include 'agents/partials/common_agent_system_prompt.j2' %}
|
|
6
6
|
|
|
7
|
+
## MEMORY MANAGEMENT PROTOCOL
|
|
8
|
+
|
|
9
|
+
- You have exclusive write access to: `tasks.md`
|
|
10
|
+
- SHOULD READ `research.md`, `specification.md`, and `plan.md` for context but CANNOT write to them
|
|
11
|
+
- This is your persistent memory store - ALWAYS load it first
|
|
12
|
+
- Compress content regularly to stay within context limits
|
|
13
|
+
- Keep your file updated as you work - it's your memory across sessions
|
|
14
|
+
- Archive completed tasks to a compressed section at the bottom
|
|
15
|
+
- Consolidate similar or duplicate tasks when compressing
|
|
16
|
+
- Maintain structure: Active Tasks → Backlog → Archived (compressed)
|
|
17
|
+
|
|
18
|
+
## AI AGENT PIPELINE AWARENESS
|
|
19
|
+
|
|
20
|
+
**CRITICAL**: Your output will be consumed by AI coding agents (Claude Code, Cursor, Windsurf, etc.)
|
|
21
|
+
- These agents already know how to code - don't teach programming concepts
|
|
22
|
+
- Each task should be a specific file modification or creation
|
|
23
|
+
- Format: "In [file path], [add/modify/delete] [specific code/feature]"
|
|
24
|
+
- Include acceptance criteria as executable commands (npm test, curl endpoints)
|
|
25
|
+
- Reference specific line numbers or function names when modifying existing code
|
|
26
|
+
- Break complex features into atomic, single-file tasks when possible
|
|
27
|
+
- Include validation steps that can be automated
|
|
28
|
+
- Every task should be immediately executable by an AI agent without interpretation
|
|
29
|
+
|
|
7
30
|
## TASK MANAGEMENT WORKFLOW
|
|
8
31
|
|
|
9
32
|
For task management:
|
|
@@ -73,11 +73,11 @@ shotgun/codebase/core/parser_loader.py,sha256=LZRrDS8Sp518jIu3tQW-BxdwJ86lnsTteI
|
|
|
73
73
|
shotgun/prompts/__init__.py,sha256=RswUm0HMdfm2m2YKUwUsEdRIwoczdbI7zlucoEvHYRo,132
|
|
74
74
|
shotgun/prompts/loader.py,sha256=jy24-E02pCSmz2651aCT2NgHfRrHAGMYvKrD6gs0Er8,4424
|
|
75
75
|
shotgun/prompts/agents/__init__.py,sha256=YRIJMbzpArojNX1BP5gfxxois334z_GQga8T-xyWMbY,39
|
|
76
|
-
shotgun/prompts/agents/export.j2,sha256=
|
|
77
|
-
shotgun/prompts/agents/plan.j2,sha256=
|
|
78
|
-
shotgun/prompts/agents/research.j2,sha256=
|
|
79
|
-
shotgun/prompts/agents/specify.j2,sha256=
|
|
80
|
-
shotgun/prompts/agents/tasks.j2,sha256=
|
|
76
|
+
shotgun/prompts/agents/export.j2,sha256=qnb_V4QZycaOKDevb-9ZzAY10pEgBxbdk3kKyIRDex8,5466
|
|
77
|
+
shotgun/prompts/agents/plan.j2,sha256=npu7x1p2O2O_ONI0nmkSvG7Kv6OC1dU8_IuqTOV1sHU,3420
|
|
78
|
+
shotgun/prompts/agents/research.j2,sha256=JBtjXaMVDRuNTt7-Ai8gUb2InfolfqCkQoEkn9PsQZk,3929
|
|
79
|
+
shotgun/prompts/agents/specify.j2,sha256=AP7XrA3KE7GZsCvW4guASxZHBM2mnrMw3irdZ3RUOBs,2808
|
|
80
|
+
shotgun/prompts/agents/tasks.j2,sha256=Vm6nFSgYzdwRBySUzwMzWHhfYR12Cb_xKPqif7yMuhs,4909
|
|
81
81
|
shotgun/prompts/agents/partials/codebase_understanding.j2,sha256=7WH-PVd-TRBFQUdOdKkwwn9hAUaJznFZMAGHhO7IGGU,5633
|
|
82
82
|
shotgun/prompts/agents/partials/common_agent_system_prompt.j2,sha256=f5cpo2Ds7EceHvXYUHNs_1NoXv_K4BrWduXYcT2PL40,1803
|
|
83
83
|
shotgun/prompts/agents/partials/content_formatting.j2,sha256=MG0JB7SSp8YV5akDWpbs2f9DcdREIYqLp38NnoWLeQ0,1854
|
|
@@ -121,8 +121,8 @@ shotgun/utils/__init__.py,sha256=WinIEp9oL2iMrWaDkXz2QX4nYVPAm8C9aBSKTeEwLtE,198
|
|
|
121
121
|
shotgun/utils/env_utils.py,sha256=8QK5aw_f_V2AVTleQQlcL0RnD4sPJWXlDG46fsHu0d8,1057
|
|
122
122
|
shotgun/utils/file_system_utils.py,sha256=l-0p1bEHF34OU19MahnRFdClHufThfGAjQ431teAIp0,1004
|
|
123
123
|
shotgun/utils/update_checker.py,sha256=Xf-7w3Pos3etzCoT771gJe2HLkA8_V2GrqWy7ni9UqA,11373
|
|
124
|
-
shotgun_sh-0.1.0.
|
|
125
|
-
shotgun_sh-0.1.0.
|
|
126
|
-
shotgun_sh-0.1.0.
|
|
127
|
-
shotgun_sh-0.1.0.
|
|
128
|
-
shotgun_sh-0.1.0.
|
|
124
|
+
shotgun_sh-0.1.0.dev25.dist-info/METADATA,sha256=GkDNaK-wokG1FYKaeQOLGZ2vVS0AehWd82rYBpjmbes,11197
|
|
125
|
+
shotgun_sh-0.1.0.dev25.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
|
|
126
|
+
shotgun_sh-0.1.0.dev25.dist-info/entry_points.txt,sha256=asZxLU4QILneq0MWW10saVCZc4VWhZfb0wFZvERnzfA,45
|
|
127
|
+
shotgun_sh-0.1.0.dev25.dist-info/licenses/LICENSE,sha256=YebsZl590zCHrF_acCU5pmNt0pnAfD2DmAnevJPB1tY,1065
|
|
128
|
+
shotgun_sh-0.1.0.dev25.dist-info/RECORD,,
|
|
File without changes
|
|
File without changes
|
|
File without changes
|